CN109077719A - signal identification method, device, equipment and storage medium - Google Patents
signal identification method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN109077719A CN109077719A CN201810729612.2A CN201810729612A CN109077719A CN 109077719 A CN109077719 A CN 109077719A CN 201810729612 A CN201810729612 A CN 201810729612A CN 109077719 A CN109077719 A CN 109077719A
- Authority
- CN
- China
- Prior art keywords
- electrocardiosignal
- feature parameter
- signal
- parameter
- vector machine
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 52
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 21
- 238000000605 extraction Methods 0.000 claims abstract description 21
- 238000012549 training Methods 0.000 claims abstract description 17
- 238000012706 support-vector machine Methods 0.000 claims abstract description 9
- 230000016507 interphase Effects 0.000 claims description 41
- 239000000284 extract Substances 0.000 claims description 19
- 238000004590 computer program Methods 0.000 claims description 4
- 238000007781 pre-processing Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 2
- 206010003658 Atrial Fibrillation Diseases 0.000 description 18
- 230000033764 rhythmic process Effects 0.000 description 14
- 238000012545 processing Methods 0.000 description 11
- 230000008569 process Effects 0.000 description 10
- 238000010586 diagram Methods 0.000 description 9
- 238000001514 detection method Methods 0.000 description 5
- 239000006185 dispersion Substances 0.000 description 5
- 230000005291 magnetic effect Effects 0.000 description 5
- 230000002159 abnormal effect Effects 0.000 description 4
- 238000004891 communication Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 4
- 238000013461 design Methods 0.000 description 3
- 238000003745 diagnosis Methods 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 230000007306 turnover Effects 0.000 description 3
- 206010003119 arrhythmia Diseases 0.000 description 2
- 230000001746 atrial effect Effects 0.000 description 2
- 230000000747 cardiac effect Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000002790 cross-validation Methods 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000005611 electricity Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000028161 membrane depolarization Effects 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 230000002861 ventricular Effects 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 230000003321 amplification Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 230000002060 circadian Effects 0.000 description 1
- 238000010224 classification analysis Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000035475 disorder Diseases 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 210000005246 left atrium Anatomy 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000002107 myocardial effect Effects 0.000 description 1
- 208000010125 myocardial infarction Diseases 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 230000004213 regulation of atrial cardiomyocyte membrane depolarization Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000000192 social effect Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
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
- A61B5/361—Detecting fibrillation
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Cardiology (AREA)
- Heart & Thoracic Surgery (AREA)
- Molecular Biology (AREA)
- Pathology (AREA)
- Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- Biophysics (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The embodiment of the invention discloses a signal identification method, a device, equipment and a storage medium, wherein the method comprises the following steps: extracting a first characteristic parameter of the electrocardiosignal, wherein the first characteristic parameter is obtained by extraction according to waveform information of the electrocardiosignal, a wave sequence of the electrocardiosignal and a set operator; extracting a second characteristic parameter of the electrocardiosignal, wherein the second characteristic parameter is obtained by extraction according to a set turning point curvature algorithm; taking the first characteristic parameter and the second characteristic parameter as input samples of a support vector machine model to obtain model parameters of the support vector machine model; and inputting the signal to be detected into the support vector machine model for training, and identifying the signal to be detected. The method in the embodiment is applied to signal identification, and the speed and the accuracy of signal identification are improved.
Description
Technical field
The present invention relates to signal processing technology more particularly to a kind of signal recognition method, device, equipment and storage mediums.
Background technique
Auricular fibrillation abbreviation atrial fibrillation is clinical one of the most common type cardiac arrhythmia, reduces the disease incidence of atrial fibrillation and dead
Rate is died with important clinical meaning and social effect.
In the implementation of the present invention, at least there are the following problems in the prior art for inventor's discovery.In the prior art
Standard is used as irregularly by RR interphase to judge whether atrial fibrillation breaks out, however, RR interphase is absolutely irregularly also other rhythms of the heart
One of performance of signal.Judgment criteria is single to cause judging result inaccurate.
Summary of the invention
The embodiment of the present invention provides a kind of signal recognition method, device, equipment and storage medium, using in the present embodiment
Method carries out signal identification, improves the speed and accuracy of signal identification.
In a first aspect, the embodiment of the invention provides a kind of signal recognition methods, this method comprises:
Extract the fisrt feature parameter of electrocardiosignal, wherein the fisrt feature parameter is the waveform according to electrocardiosignal
Information, the wave train of electrocardiosignal and setting operation operator extraction obtain；
Extract the second feature parameter of electrocardiosignal, wherein the second feature parameter is bent according to the turning point of setting
Rate algorithm, which extracts, to be obtained；
Using the fisrt feature parameter and the second feature parameter as the input sample of supporting vector machine model, obtain
The model parameter of the supporting vector machine model；
Signal to be detected is inputted the supporting vector machine model to be trained, identifies signal to be detected.
Second aspect, the embodiment of the invention also provides a kind of signal recognition device, which includes:
First extraction module, for extracting the fisrt feature parameter of electrocardiosignal, wherein the fisrt feature parameter is root
It is obtained according to the shape information of electrocardiosignal, the wave train of electrocardiosignal and setting operation operator extraction；
Second extraction module, for extracting the second feature parameter of electrocardiosignal, wherein the second feature parameter is root
It extracts and obtains according to the turning point curvature algorithm of setting；
Model parameter obtains module, for using the fisrt feature parameter and the second feature parameter as supporting vector
The input sample of machine model obtains the model parameter of the supporting vector machine model；
Signal identification module is trained for signal to be detected to be inputted the supporting vector machine model, identifies to be checked
Survey signal.
The third aspect the embodiment of the invention also provides a kind of computer equipment, including memory, processor and is stored in
On memory and the computer program that can run on a processor, the processor are realized when executing described program as the present invention is real
Apply any signal recognition method in example.
Fourth aspect, the embodiment of the invention also provides a kind of computer readable storage mediums, are stored thereon with computer
Program realizes the signal recognition method as described in any in the embodiment of the present invention when program is executed by processor.
In the embodiment of the present invention, by extracting the fisrt feature parameter of electrocardiosignal, fisrt feature parameter is according to electrocardio
The shape information of signal, the wave train of electrocardiosignal and setting operation operator extraction obtain, and extract electrocardiosignal second is special
Parameter is levied, second feature parameter is to extract to obtain according to the turning point curvature algorithm of setting；By the fisrt feature parameter and institute
Input sample of the second feature parameter as supporting vector machine model is stated, the model parameter of the supporting vector machine model is obtained；
Signal to be detected is inputted the supporting vector machine model to be trained, identifies signal to be detected.Using in the embodiment of the present invention
The method of offer carries out signal identification, improves the speed and accuracy of signal identification；It is more smooth by pretreated waveform, in advance
Processing eliminates high-frequency noise, is more convenient accurately to extract effective information；Using the characteristic parameter provided in the embodiment of the present invention,
More accurate model parameter value can be obtained, signal to be detected is identified using supporting vector machine model, is better able to characterize
The characteristic of atrial attack, is more suitable for practical application scene；The result is shown on the display device, as personal or doctor detection
Or the basis of diagnosis.
Detailed description of the invention
Fig. 1 is the flow chart of one of the embodiment of the present invention one signal recognition method；
Fig. 2 a is the flow chart of one of the embodiment of the present invention two signal recognition method；
Fig. 2 b is the waveform diagram for the electrocardiosignal that a kind of actual acquisition being applicable in the embodiment of the present invention two arrives；
Fig. 2 c is a kind of waveform diagram for the pretreated electrocardiosignal being applicable in the embodiment of the present invention two；
Fig. 3 is the flow chart of one of the embodiment of the present invention three signal recognition method；
Fig. 4 is the structural schematic diagram of one of the embodiment of the present invention four signal recognition device；
Fig. 5 is the structural schematic diagram of one of the embodiment of the present invention five computer equipment.
Specific embodiment
The present invention is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining the present invention rather than limiting the invention.It also should be noted that in order to just
Only the parts related to the present invention are shown in description, attached drawing rather than entire infrastructure.
Electrocardiosignal (Electrocardiogram, ECG) refers to, the biological telecommunications that myocardial activation generates when cardiomotility
Number.By identify electrocardiosignal, can identify atrial fibrillation, other rhythms of the heart (abnormal rhythm in addition to atrial fibrillation), normal sinus rhythm and
The different electrocardiosignal such as noise.Atrial fibrillation refers to auricular fibrillation, is clinical one of the most common type cardiac arrhythmia, the heart of disorder
The complication such as room activity and the following brain soldier, myocardial infarction, lead to higher disability rate and the death rate.Therefore, accurate to know
Other electrocardiosignal has important meaning.
In a specific example, electrocardiosignal is shown on electrocardiograph, the electrocardiosignal shown on electrocardiograph
By a series of wave component, each wave group represents a cardiac cycle, and a wave group includes P wave, QRS wave and T wave.P wave is
Atrial depolarization wave is the first wave in each wave group, reflects the process of depolarization of left atrium；Typical QRS wave includes three
A closely coupled wave, first downward wave are known as Q wave, and the upright wave of the high point of after Q wave is known as R wave, after R wave
Downward wave is known as S wave, because its is closely coupled, and reflects ventricle electricity ignition process, therefore is referred to as QRS wave, reflects left and right
The process of depolarization of ventricle.T wave is located at after S-T segment, is that one relatively low and while accounting for longer wave, it is produced by ventricular bipolar
's.U wave is the low frequency that immediately 0.02~0.04s occurs after T wave, short arc wave, and direction is consistent with T wave, is ventricular bipolar
A part.
Embodiment one
Fig. 1 is a kind of flow chart for signal recognition method that the embodiment of the present invention one provides, and the present embodiment is applicable to know
Not signal to be detected the case where, this method can be executed by signal recognition device provided in an embodiment of the present invention, which can
It is realized by the way of software and/or hardware.With reference to Fig. 1, this method can specifically include following steps:
S110, the fisrt feature parameter for extracting electrocardiosignal, wherein the fisrt feature parameter is according to electrocardiosignal
Shape information, the wave train of electrocardiosignal and setting operation operator extraction obtain.
Specifically, extracting the fisrt feature parameter of electrocardiosignal, wherein the statistics that fisrt feature parameter can be ECG is special
Sign.Optionally, the fisrt feature parameter include: the quantity of electrocardiosignal wave, average value, maximum value, minimum value, median and
Variance.
Fisrt feature is determined according to the operation operator of the shape information of electrocardiosignal, the wave train of electrocardiosignal and setting
Parameter.Wherein, the shape information of electrocardiosignal can be crest location, wave trough position and wave interphase.Optionally, wave interphase is
Refer to, the time difference between the different wave crests of the same waveform or the time difference between the wave crest of different wave, is not formed here
It is specific to limit, depending on actual applicable cases.The wave train of electrocardiosignal can be the discrete of multiple electrocardiosignal compositions
The operation operator of sequence, setting can be sum, Mean, Max, Min, Median and Var etc..
In a specific example, with enabling X_P, X_Q, X_R, X_S and X_T indicate P wave, Q wave, R wave, S wave and T wave
Sequence, by taking R wave as an example, calculate the quantity Num of R wave, average value Mean, maximum value Max, minimum M in, median Median and
Variance Var, wherein average value, maximum value, minimum value, median and variance are for signal amplitude, and amplitude can be
Voltage.Illustrate the calculating process of fisrt feature parameter with a specific example.
Num_R=Num (X_R)=sum (X_R)
Mean_R=Mean (X_R)
Max_R=Max (X_R)
Min_R=Min (X_R)
Median_R=Median (X_R)
Var_R=Var (X_R)
Similarly, the fisrt feature parameter of the acquisition process and P wave of the fisrt feature parameter of Q wave, R wave, S wave and T wave obtains
The process of obtaining is identical, and this will not be repeated here.
S120, the second feature parameter for extracting electrocardiosignal, wherein the second feature parameter is the turnover according to setting
Point curvature algorithm, which extracts, to be obtained.
Wherein, by taking R wave as an example, second feature parameter may include the dispersion characteristic parameter of RR interphase difference △ RR, according to
Dispersion calculation of characteristic parameters is set in the sequence of the △ RR interphase random alignment of length, the confidence occurred by calculating turning point
Section determines second feature parameter.Similarly, the acquisition process and P of P wave, Q wave, R wave, S wave and the relevant second feature parameter of T wave
The acquisition process of the second feature parameter of wave is identical, and this will not be repeated here.
S130, using the fisrt feature parameter and the second feature parameter as the input sample of supporting vector machine model
This, obtains the model parameter of the supporting vector machine model.
In a specific example, using fisrt feature parameter and second feature parameter as support vector machines (Support
Vector Machine, SVM) input sample x, by " atrial fibrillation ", " other abnormal rhythms ", " regular sinus rhythm " and " noise "
Labeled as the output y of SVM, (x, y) collectively constitutes the training sample pair of SVM, carries out SVM training, obtains the model parameter of SVM.
S140, the signal to be detected input supporting vector machine model is trained, identifies signal to be detected.
Specifically, signal to be detected input supporting vector machine model is carried out using the supporting vector machine model got
Training identifies signal to be detected according to training result.
Optionally, described to be trained the signal to be detected input supporting vector machine model, identify signal to be detected
Later, further includes: show recognition result on the display device, wherein shown recognition result includes signal type.Wherein, know
Other result can be atrial fibrillation signal, other rhythm of the heart (abnormal rhythm in addition to atrial fibrillation), normal sinus rhythm and noises these four not
Same electrocardiosignal.Using the method provided in the embodiment of the present invention, it is better able to the characteristic of characterization atrial attack, is more suitable for reality
Border application scenarios.
The result is shown on the display device, the basis as personal or doctor detection or diagnosis.It is specific at one
In example, display equipment can be the patch of single lead electrocardio comprising ECG module, more sign devices or patient monitor equipment etc..
Technical solution provided in an embodiment of the present invention provides good selection for detection atrial fibrillation, may operate in portable
On electrocardiograph or atrial fibrillation detection device.By the target classification of subdivision atrial fibrillation detection, simple atrial fibrillation, non-atrial fibrillation are classified as
More careful atrial fibrillation, other abnormal rhythms, regular sinus rhythm and noise can more accurately assist doctor to carry out profession
Judgement, it is also more practical.
In the embodiment of the present invention, by extracting the fisrt feature parameter of electrocardiosignal, fisrt feature parameter is according to electrocardio
The shape information of signal, the wave train of electrocardiosignal and setting operation operator extraction obtain, and extract electrocardiosignal second is special
Parameter is levied, second feature parameter is to extract to obtain according to the turning point curvature algorithm of setting；By the fisrt feature parameter and institute
Input sample of the second feature parameter as supporting vector machine model is stated, the model parameter of the supporting vector machine model is obtained；
Signal to be detected is inputted the supporting vector machine model to be trained, identifies signal to be detected.Using the side in the present embodiment
Method carries out signal identification, improves the speed and accuracy of signal identification.
Embodiment two
Fig. 2 a is a kind of flow chart of signal recognition method provided by Embodiment 2 of the present invention, and the present embodiment is in above-mentioned implementation
It is realized on the basis of example.With reference to Fig. 2 a, this method can specifically include following steps:
S210, it obtains electrocardiosignal and is pre-processed.
Specifically, being acquired by dedicated signal collecting device to electrocardiosignal, dedicated signal collecting device can
To be electrocardiograph, the signal after acquisition is pre-processed.In a specific example, adopted using multi-channel synchronous data
Human heart signal, ambient noise and the electrocardiosignal of processing are acquired and are stored by collection.Firstly, passing through cardiac diagnosis lead and biography
Sensor obtains electrocardiogram (ECG) data, carries out the processing such as impedance matching, filtering and amplification by signal of the analog circuit to acquisition.Then,
The analog signal of human body physiological parameter is converted into digital signal by analog-digital converter, is stored by memory.Fig. 2 b shows one
The waveform diagram for the electrocardiosignal that kind actual acquisition arrives, wherein it include various noises in electrocardiosignal, waveform is coarse, and it is rough, it leads
The useful information contained in QRS wave is caused to be difficult to extract.Low pass filtered is carried out using lowpass digital filter (Butterworth filter)
Wave filters out high-frequency noise (300Hz or more), obtains filtered electrocardiosignal.
S220, the shape information for determining treated electrocardiosignal.
Illustratively, the shape information of the P wave and QRS wave in electrocardiosignal is extracted using wavelet transformation technique, optionally,
Shape information includes PR interphase, RR interphase and QT interphase.The datum mark of electrocardiosignal, i.e. P wave, Q are obtained by TP and PQ baseline
The position of the wave crest of wave, R wave, S wave and T wave, and PR interphase is calculated, RR interphase, QT interphase etc..Optionally, interphase refers to
Time interval between two wave crests, for example, PR interphase refers to, the time interval between the wave crest of P wave and the wave crest of R wave.Figure
2c shows a kind of waveform diagram of pretreated electrocardiosignal, can be seen that by Fig. 2 c by pretreated waveform more light
It is sliding.
S230, the fisrt feature parameter for extracting electrocardiosignal, wherein the fisrt feature parameter is according to electrocardiosignal
Shape information, the wave train of electrocardiosignal and setting operation operator extraction obtain.
S240, the second feature parameter for extracting electrocardiosignal, wherein the second feature parameter is the turnover according to setting
Point curvature algorithm, which extracts, to be obtained.
S250, using the fisrt feature parameter and the second feature parameter as the input sample of supporting vector machine model
This, obtains the model parameter of the supporting vector machine model.
S260, the signal to be detected input supporting vector machine model is trained, identifies signal to be detected.
In the embodiment of the present invention, before the fisrt feature parameter for extracting electrocardiosignal, acquisition electrocardiosignal first is gone forward side by side
Row pretreatment, determines the shape information of treated electrocardiosignal.High-frequency noise is eliminated, effective information is more accurately extracted.
Illustratively, the second feature parameter extracted in electrocardiosignal can be accomplished in that according to setting
Spaced discrete degree feature during turning point curvature algorithm extracts the spaced discrete degree characteristic parameter of the RR interphase, the PR is joined
Several and the QT interphase spaced discrete degree characteristic parameter.
The first, the spaced discrete degree characteristic parameter of RR interphase is extracted according to the turning point curvature algorithm of setting.
In a specific example, design turning point curvature algorithm extracts the dispersion feature ginseng of RR interphase difference △ RR
Number.Assuming that three adjacent △ RR interphases are a1, a2, a3, a1 > a2 > a3, then these three △ RR interphases have 6 kinds of arrangement modes
As follows, (a1, a3, a2), (a2, a1, a3), (a3, a1, a2), this 4 kinds of (a2, a3, a1) are turning point, (a1, a2, a3), (a3,
A2, a1) this 2 kinds be non-turning point, i.e., turning point occur probability be 2/3.So, a length be l △ RR interphase with
In the sequence of machine arrangement, calculating the confidence interval that turning point occurs is TPR_RR, wherein mean value is (2l-4)/3, and standard deviation isConfidence interval
The second, the spaced discrete degree characteristic parameter during extracting PR according to the turning point curvature algorithm of setting.
Illustratively, design turning point curvature algorithm extracts the dispersion feature of PR interphase difference △ PR.Assuming that three adjacent
△ PR interphase be b1, b2, b3, b1 > b2 > b3, then these three △ PR interphases have 6 kinds of arrangement modes as follows, (b1, b3, b2),
(b2, b1, b3), (b3, b1, b2), this 4 kinds of (b2, b3, b1) are turning point, this 2 kinds of (b1, b2, b3), (b3, b2, b1) is non-
The probability that turning point, i.e. turning point occur is 2/3.So, in the sequence for the △ PR interphase random alignment that a length is l,
Calculating the confidence interval that turning point occurs is TPR_PR, wherein mean value is (2l-4)/3, and standard deviation isConfidence
Section
Third, the spaced discrete degree characteristic parameter that QT interphase is extracted according to the turning point curvature algorithm of setting.
In a specific example, design turning point curvature algorithm extracts the dispersion feature of QT interphase difference △ QT, false
If three adjacent △ QT interphases are c1, c2, c3, c1 > c2 > c3, then these three △ QT interphases have 6 kinds of arrangement modes as follows,
(c1, c3, c2), (c2, c1, c3), (c3, c1, c2), this 4 kinds of (c2, c3, c1) are turning point, (c1, c2, c3), (c3, c2,
C1) this 2 kinds are non-turning point, i.e., the probability that turning point occurs is 2/3.So, it is arranged at random in the △ QT interphase that a length is l
In the sequence of column, calculating the confidence interval that turning point occurs is TPR_QT, wherein mean value is (2l-4)/3, and standard deviation isConfidence interval
By spaced discrete degree characteristic parameter during the spaced discrete degree characteristic parameter of above-mentioned RR interphase, the PR and described
Input sample of the spaced discrete degree characteristic parameter of QT interphase as supporting vector machine model improves the standard that model parameter determines
True property.It should be noted that the calculating of above-mentioned spaced discrete degree characteristic parameter is used only to example, not to technology of the invention
Scheme, which is formed, to be limited.
Embodiment three
Fig. 3 is a kind of flow chart for signal recognition method that the embodiment of the present invention three provides, and the present embodiment is in above-mentioned implementation
It is realized on the basis of example.With reference to Fig. 3, this method can specifically include following steps:
S310, the fisrt feature parameter for extracting electrocardiosignal, wherein the fisrt feature parameter is according to electrocardiosignal
Shape information, the wave train of electrocardiosignal and setting operation operator extraction obtain.
S320, the second feature parameter for extracting electrocardiosignal, wherein the second feature parameter is the turnover according to setting
Point curvature algorithm, which extracts, to be obtained.
S330, using the fisrt feature parameter and the second feature parameter as the input sample of supporting vector machine model
This, obtains the model parameter of the supporting vector machine model.
S340, the supporting vector machine model is determined according to the model parameter of the supporting vector machine model.
Wherein, supporting vector machine model is supervised learning model related to relevant learning algorithm, can analyze data
Mode and recognition mode, for classification and regression analysis.The model determination process of support vector machines in the embodiment of the present invention is such as
Under: fisrt feature parameter and second feature parameter are extracted according to the electrocardiosignal of input, svm classifier mould is established by training sample
Type.
Firstly, for given sample to { (xi, yi), xi ∈ RN, yi={ 0,1,2,3 } }, xi is training sample, RNFor sample
This number.During the determination of supporting vector machine model parameter, xi is to the fisrt feature determined after original electro-cardiologic signals processing
The training sample that parameter and second feature parameter are used as；Yi sample to be adjudicated, that is, for the rhythm signal type of setting, such as
Y1 is atrial fibrillation signal, y2 is other circadian signals, y3 is normal sinus rhythm, y4 is the rhythm of the heart.Next to support vector machines mould
Common Parameters in type training process are introduced.
Step 1: C is arranged in the section [C1, C2], i.e. C ∈ [C1, C2], change step Cs, and γ setting exists
[γ1, γ2] in section, i.e. γ ∈ [γ1, γ2], change step γs.Example C ∈ [2-10, 210], Cs=2；γ∈[2-10,
210], γs=2.It is trained for each pair of parameter (C, γ), a pair of of the parameter for taking effect best is as model parameter.
Step 2: for the various combination of fisrt feature parameter and second feature parameter, training dataset being divided into k parts of phases
Deng subset, every time will wherein k-1 parts of data as training data, and will in addition a data as test data.Repeat k
It is secondary, estimate it is expected extensive error according to the mean square error average value obtained after k iteration, and obtain cross validation accuracy.
Step 3: according to aforementioned fisrt feature parameter and the further subdivided meshes of second feature parameter area, obtaining more accurate
Parameter value, according to cross validation average accuracy sort, the highest parameter combination of selection sort accuracy as model most
Excellent parameter.
Step 4: data being divided into training set and test set again, using the most optimized parameter model training model, utilize survey
Try data test model performance.
Finally, the optimized parameter based on training sample to the model obtained with training, obtains trained SVM model.
S350, the signal to be detected input supporting vector machine model is trained, identifies signal to be detected.
In the embodiment of the present invention, by the way that fisrt feature parameter and second feature parameter are determined that supporting vector machine model is joined
Number, can obtain more accurate model parameter value.
Example IV
Fig. 4 is a kind of structural schematic diagram for signal recognition device that the embodiment of the present invention four provides, which is suitable for holding
A kind of signal recognition method that the row embodiment of the present invention is supplied to.As shown in figure 4, the device can specifically include:
First extraction module 410, for extracting the fisrt feature parameter of electrocardiosignal, wherein the fisrt feature parameter
To be obtained according to the shape information of electrocardiosignal, the wave train of electrocardiosignal and setting operation operator extraction；
Second extraction module 420, for extracting the second feature parameter of electrocardiosignal, wherein the second feature parameter
It is obtained to be extracted according to the turning point curvature algorithm of setting；
Model parameter obtains module 430, for using the fisrt feature parameter and the second feature parameter as support
The input sample of vector machine model obtains the model parameter of the supporting vector machine model；
Signal identification module 440 is trained for signal to be detected to be inputted the supporting vector machine model, identification to
Detect signal.
Further, further includes:
Preprocessing module, for before the fisrt feature parameter for extracting electrocardiosignal, obtaining electrocardiosignal and carrying out pre-
Processing；
Information determination module, for determining the shape information of treated electrocardiosignal.
Further, the shape information includes PR interphase, RR interphase and QT interphase；
Correspondingly, the second extraction module 420 is specifically used for:
During extracting the spaced discrete degree characteristic parameter of the RR interphase, the PR according to the turning point curvature algorithm of setting
Spaced discrete degree characteristic parameter and the QT interphase spaced discrete degree characteristic parameter.
It further, further include model determining module, for signal to be detected to be inputted the supporting vector machine model
It is trained, before identifying signal to be detected, the supporting vector is determined according to the model parameter of the supporting vector machine model
Machine model.
Further, the fisrt feature parameter include: the quantity of electrocardiosignal wave, average value, maximum value, minimum value,
Median and variance.
It further, further include display module, for signal to be detected to be inputted the supporting vector machine model progress
Training after identifying signal to be detected, recognition result is shown on the display device, wherein shown recognition result includes signal
Type.
The signal identification side that any embodiment of that present invention provides can be performed in signal recognition device provided in an embodiment of the present invention
Method has the corresponding functional module of execution method and beneficial effect.
Embodiment five
Fig. 5 is a kind of structural schematic diagram for computer equipment that the embodiment of the present invention five provides.Fig. 5, which is shown, to be suitable for being used to
Realize the block diagram of the exemplary computer device 12 of embodiment of the present invention.The computer equipment 12 that Fig. 5 is shown is only one
Example, should not function to the embodiment of the present invention and use scope bring any restrictions.
As shown in figure 5, computer equipment 12 is showed in the form of universal computing device.The component of computer equipment 12 can be with
Including but not limited to: one or more processor or processing unit 16, system storage 28 connect different system components
The bus 18 of (including system storage 28 and processing unit 16).
Bus 18 indicates one of a few class bus structures or a variety of, including memory bus or Memory Controller,
Peripheral bus, graphics acceleration port, processor or the local bus using any bus structures in a variety of bus structures.It lifts
For example, these architectures include but is not limited to industry standard architecture (ISA) bus, microchannel architecture (MAC)
Bus, enhanced isa bus, Video Electronics Standards Association (VESA) local bus and peripheral component interconnection (PCI) bus.
Computer equipment 12 typically comprises a variety of computer system readable media.These media can be it is any can be by
The usable medium that computer equipment 12 accesses, including volatile and non-volatile media, moveable and immovable medium.
System storage 28 may include the computer system readable media of form of volatile memory, such as arbitrary access
Memory (RAM) 30 and/or cache memory 32.Computer equipment 12 may further include it is other it is removable/can not
Mobile, volatile/non-volatile computer system storage medium.Only as an example, storage system 34 can be used for reading and writing not
Movably, non-volatile magnetic media (Fig. 5 do not show, commonly referred to as " hard disk drive ").It, can be with although being not shown in Fig. 5
The disc driver for reading and writing to removable non-volatile magnetic disk (such as " floppy disk ") is provided, and non-volatile to moving
The CD drive of CD (such as CD-ROM, DVD-ROM or other optical mediums) read-write.In these cases, each driving
Device can be connected by one or more data media interfaces with bus 18.System storage 28 may include at least one journey
Sequence product, the program product have one group of (for example, at least one) program module, these program modules are configured to perform this hair
The function of bright each embodiment.
Program/utility 40 with one group of (at least one) program module 42 can store and store in such as system
In device 28, such program module 42 includes --- but being not limited to --- operating system, one or more application program, other
It may include the realization of network environment in program module and program data, each of these examples or certain combination.Journey
Sequence module 42 usually executes function and/or method in embodiment described in the invention.
Computer equipment 12 can also be with one or more external equipments 14 (such as keyboard, sensing equipment, display 24
Deng) communication, can also be enabled a user to one or more equipment interact with the computer equipment 12 communicate, and/or with make
The computer equipment 12 any equipment (such as network interface card, the modulatedemodulate that can be communicated with one or more of the other calculating equipment
Adjust device etc.) communication.This communication can be carried out by input/output (I/O) interface 22.Also, computer equipment 12 may be used also
To pass through network adapter 20 and one or more network (such as local area network (LAN), wide area network (WAN) and/or public network
Network, such as internet) communication.As shown, network adapter 20 is logical by other modules of bus 18 and computer equipment 12
Letter.It should be understood that other hardware and/or software module, packet can be used in conjunction with computer equipment 12 although being not shown in Fig. 5
It includes but is not limited to: microcode, device driver, redundant processing unit, external disk drive array, RAID system, magnetic tape drive
Device and data backup storage system etc..
Processing unit 16 by the program that is stored in system storage 28 of operation, thereby executing various function application and
Data processing, such as realize signal recognition method provided by the embodiment of the present invention:
That is, the processing unit is realized when executing described program: extracting the fisrt feature parameter of electrocardiosignal, wherein
The fisrt feature parameter is according to the shape information of electrocardiosignal, the wave train of electrocardiosignal and setting operation operator extraction
It obtains；Extract the second feature parameter of electrocardiosignal, wherein the second feature parameter is to calculate according to the turning point curvature of setting
Method, which is extracted, to be obtained；Using the fisrt feature parameter and the second feature parameter as the input sample of supporting vector machine model,
Obtain the model parameter of the supporting vector machine model；Signal to be detected is inputted the supporting vector machine model to be trained,
Identify signal to be detected.
Embodiment six
The embodiment of the present invention six provides a kind of computer readable storage medium, is stored thereon with computer program, the journey
The signal recognition method provided such as all inventive embodiments of the application is provided when sequence is executed by processor:
That is, realization when the program is executed by processor: extracting the fisrt feature parameter of electrocardiosignal, wherein described
One characteristic parameter is to be obtained according to the shape information of electrocardiosignal, the wave train of electrocardiosignal and setting operation operator extraction；
Extract the second feature parameter of electrocardiosignal, wherein the second feature parameter is to mention according to the turning point curvature algorithm of setting
Take acquisition；Using the fisrt feature parameter and the second feature parameter as the input sample of supporting vector machine model, obtain
The model parameter of the supporting vector machine model；Signal to be detected is inputted the supporting vector machine model to be trained, is identified
Signal to be detected.
It can be using any combination of one or more computer-readable media.Computer-readable medium can be calculating
Machine readable signal medium or computer readable storage medium.Computer readable storage medium for example can be --- but it is unlimited
In system, device or the device of --- electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor, or any above combination.It calculates
The more specific example (non exhaustive list) of machine readable storage medium storing program for executing includes: electrical connection with one or more conducting wires, just
Taking formula computer disk, hard disk, random access memory (RAM), read-only memory (ROM), erasable type may be programmed read-only storage
Device (EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), light storage device, magnetic memory device,
Or above-mentioned any appropriate combination.In this document, computer readable storage medium can be it is any include or storage journey
The tangible medium of sequence, the program can be commanded execution system, device or device use or in connection.
Computer-readable signal media may include in a base band or as carrier wave a part propagate data-signal,
Wherein carry computer-readable program code.The data-signal of this propagation can take various forms, including --- but
It is not limited to --- electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be
Any computer-readable medium other than computer readable storage medium, which can send, propagate or
Transmission is for by the use of instruction execution system, device or device or program in connection.
The program code for including on computer-readable medium can transmit with any suitable medium, including --- but it is unlimited
In --- wireless, electric wire, optical cable, RF etc. or above-mentioned any appropriate combination.
The computer for executing operation of the present invention can be write with one or more programming languages or combinations thereof
Program code, described program design language include object oriented program language-such as Java, Smalltalk, C++,
It further include conventional procedural programming language-such as " C " language or similar programming language.Program code can be with
It fully executes, partly execute on the user computer on the user computer, being executed as an independent software package, portion
Divide and partially executes or executed on a remote computer or server completely on the remote computer on the user computer.?
Be related in the situation of remote computer, remote computer can pass through the network of any kind --- including local area network (LAN) or
Wide area network (WAN)-be connected to subscriber computer, or, it may be connected to outer computer (such as mentioned using Internet service
It is connected for quotient by internet).
Note that the above is only a better embodiment of the present invention and the applied technical principle.It will be appreciated by those skilled in the art that
The invention is not limited to the specific embodiments described herein, be able to carry out for a person skilled in the art it is various it is apparent variation,
It readjusts and substitutes without departing from protection scope of the present invention.Therefore, although being carried out by above embodiments to the present invention
It is described in further detail, but the present invention is not limited to the above embodiments only, without departing from the inventive concept, also
It may include more other equivalent embodiments, and the scope of the invention is determined by the scope of the appended claims.
Claims (10)
1. a kind of signal recognition method characterized by comprising
Extract the fisrt feature parameter of electrocardiosignal, wherein the fisrt feature parameter be according to the shape information of electrocardiosignal,
The wave train and setting operation operator extraction of electrocardiosignal obtain；
Extract the second feature parameter of electrocardiosignal, wherein the second feature parameter is to calculate according to the turning point curvature of setting
Method, which is extracted, to be obtained；
Using the fisrt feature parameter and the second feature parameter as the input sample of supporting vector machine model, described in acquisition
The model parameter of supporting vector machine model；
Signal to be detected is inputted the supporting vector machine model to be trained, identifies signal to be detected.
2. the method according to claim 1, wherein it is described extract electrocardiosignal fisrt feature parameter before,
Further include:
It obtains electrocardiosignal and is pre-processed；
Determine the shape information of treated electrocardiosignal.
3. according to the method described in claim 2, it is characterized in that, the shape information includes between PR interphase, RR interphase and QT
Phase；
Correspondingly, the second feature parameter extracted in electrocardiosignal, comprising:
Between during extracting the spaced discrete degree characteristic parameter of the RR interphase, the PR according to the turning point curvature algorithm of setting
The spaced discrete degree characteristic parameter of divergence characteristic parameter and the QT interphase is isolated.
4. the method according to claim 1, wherein by signal to be detected input the supporting vector machine model into
Row training, before identifying signal to be detected, further includes:
The supporting vector machine model is determined according to the model parameter of the supporting vector machine model.
5. the method according to claim 1, wherein the fisrt feature parameter includes: the number of electrocardiosignal wave
Amount, average value, maximum value, minimum value, median and variance.
6. the method according to claim 1, wherein described input the support vector machines mould for signal to be detected
Type is trained, after identifying signal to be detected, further includes:
Recognition result is shown on the display device, wherein shown recognition result includes signal type.
7. a kind of signal recognition device characterized by comprising
First extraction module, for extracting the fisrt feature parameter of electrocardiosignal, wherein the fisrt feature parameter is according to the heart
The shape information of electric signal, the wave train of electrocardiosignal and setting operation operator extraction obtain；
Second extraction module, for extracting the second feature parameter of electrocardiosignal, wherein set according to the second feature parameter
Fixed turning point curvature algorithm, which extracts, to be obtained；
Model parameter obtains module, for using the fisrt feature parameter and the second feature parameter as support vector machines mould
The input sample of type obtains the model parameter of the supporting vector machine model；
Signal identification module is trained for signal to be detected to be inputted the supporting vector machine model, identifies letter to be detected
Number.
8. device according to claim 7, which is characterized in that further include:
Preprocessing module, for before the fisrt feature parameter for extracting electrocardiosignal, obtaining electrocardiosignal and being pre-processed；
Information determination module, for determining the shape information of treated electrocardiosignal.
9. a kind of computer equipment including memory, processor and stores the meter that can be run on a memory and on a processor
Calculation machine program, which is characterized in that the processor realizes such as side as claimed in any one of claims 1 to 6 when executing described program
Method.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is by processor
Such as method as claimed in any one of claims 1 to 6 is realized when execution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810729612.2A CN109077719A (en) | 2018-07-05 | 2018-07-05 | signal identification method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810729612.2A CN109077719A (en) | 2018-07-05 | 2018-07-05 | signal identification method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109077719A true CN109077719A (en) | 2018-12-25 |
Family
ID=64836934
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810729612.2A Pending CN109077719A (en) | 2018-07-05 | 2018-07-05 | signal identification method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109077719A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109770861A (en) * | 2019-03-29 | 2019-05-21 | 广州视源电子科技股份有限公司 | Method, device, equipment and storage medium for training and detecting cardioelectric rhythm model |
| CN110491506A (en) * | 2019-05-16 | 2019-11-22 | 中国医药大学附设医院 | Auricular fibrillation prediction model and its forecasting system |
| CN111839502A (en) * | 2020-07-21 | 2020-10-30 | 广州视源电子科技股份有限公司 | Method, device and equipment for detecting electrocardio data abnormity and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101049235A (en) * | 2007-04-26 | 2007-10-10 | 中国科学技术大学 | Body surface detection method and device for high frequency wave of repolarization of ventricle |
| US20170172513A1 (en) * | 2015-12-21 | 2017-06-22 | Industrial Technology Research Institute | Method and system for anaerobic threshold heart rate detection |
| CN107358196A (en) * | 2017-07-12 | 2017-11-17 | 北京卫嘉高科信息技术有限公司 | A kind of sorting technique of heart beat type, device and electrocardiogram equipment |
| CN107633203A (en) * | 2017-08-17 | 2018-01-26 | 平安科技（深圳）有限公司 | Facial emotions recognition methods, device and storage medium |
| CN107913062A (en) * | 2017-10-18 | 2018-04-17 | 广州视源电子科技股份有限公司 | Electrocardiosignal processing method and system |
| CN108186011A (en) * | 2017-12-13 | 2018-06-22 | 深圳竹信科技有限公司 | Atrial fibrillation detection method, device and readable storage medium storing program for executing |
-
2018
- 2018-07-05 CN CN201810729612.2A patent/CN109077719A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101049235A (en) * | 2007-04-26 | 2007-10-10 | 中国科学技术大学 | Body surface detection method and device for high frequency wave of repolarization of ventricle |
| US20170172513A1 (en) * | 2015-12-21 | 2017-06-22 | Industrial Technology Research Institute | Method and system for anaerobic threshold heart rate detection |
| CN107358196A (en) * | 2017-07-12 | 2017-11-17 | 北京卫嘉高科信息技术有限公司 | A kind of sorting technique of heart beat type, device and electrocardiogram equipment |
| CN107633203A (en) * | 2017-08-17 | 2018-01-26 | 平安科技（深圳）有限公司 | Facial emotions recognition methods, device and storage medium |
| CN107913062A (en) * | 2017-10-18 | 2018-04-17 | 广州视源电子科技股份有限公司 | Electrocardiosignal processing method and system |
| CN108186011A (en) * | 2017-12-13 | 2018-06-22 | 深圳竹信科技有限公司 | Atrial fibrillation detection method, device and readable storage medium storing program for executing |
Non-Patent Citations (3)
| Title |
|---|
| BRADLEY M WHITAKER等: ""AF classification from ECG recording using feature ensemble and sparse coding"", 《2017 COMPUTING IN CARDIOLOGY》 * |
| LUCIA BILLECI 等: ""Detection of AF and other rhythms using RR variability and ECG spectral measures"", 《2017 COMPUTING IN CARDIOLOGY》 * |
| 黄超: ""动态心电图中房颤自动检测算法研究及其临床应用"", 《中国博士学位论文数据库医药卫生科技辑》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109770861A (en) * | 2019-03-29 | 2019-05-21 | 广州视源电子科技股份有限公司 | Method, device, equipment and storage medium for training and detecting cardioelectric rhythm model |
| CN110491506A (en) * | 2019-05-16 | 2019-11-22 | 中国医药大学附设医院 | Auricular fibrillation prediction model and its forecasting system |
| CN111839502A (en) * | 2020-07-21 | 2020-10-30 | 广州视源电子科技股份有限公司 | Method, device and equipment for detecting electrocardio data abnormity and storage medium |
| CN111839502B (en) * | 2020-07-21 | 2024-02-09 | 广州视源电子科技股份有限公司 | Electrocardiogram data anomaly detection method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR102451795B1 (en) | ECG signal detection method | |
| CN109171712B (en) | Atrial fibrillation identification method, atrial fibrillation identification device, atrial fibrillation identification equipment and computer readable storage medium | |
| CN109077714A (en) | signal identification method, device, equipment and storage medium | |
| WO2021017313A1 (en) | Atrial fibrillation detection method and apparatus, computer device, and storage medium | |
| CN101828918B (en) | ECG Signal R Peak Detection Method Based on Waveform Feature Matching | |
| CN109077720A (en) | Signal processing method, device, equipment and storage medium | |
| CN111419212A (en) | Method and device for processing electrocardiogram data, storage medium and computer equipment | |
| CN110464333B (en) | Method and device for storing electrocardiogram data | |
| CN111345816B (en) | Multi-lead QRS complex detection method, device, equipment and storage medium | |
| CN106725426A (en) | A kind of method and system of electrocardiosignal classification | |
| CN109077719A (en) | signal identification method, device, equipment and storage medium | |
| CN110367969A (en) | A kind of improved electrocardiosignal K-Means Cluster | |
| CN111358459A (en) | Arrhythmia identification method, device, equipment and storage medium | |
| CN110367968B (en) | Right bundle branch retardation detection method, device, equipment and storage medium | |
| WO2024098553A1 (en) | Method and system for analyzing and identifying electrocardiogram, and storage medium | |
| CN112971800B (en) | Deep learning-based electrocardio premature beat signal processing method, module and device | |
| CN116473569A (en) | A method, system and storage medium for electrocardiogram data processing | |
| KR102387703B1 (en) | Method And Apparatus for Correcting Electrocardiogram | |
| CN118212237B (en) | Cardiovascular detection system and method based on image analysis | |
| CN104983415A (en) | Heart beat recognition method and device | |
| CN109009071A (en) | A kind of method and apparatus identifying electrocardio wave image characteristic point | |
| Lin et al. | Explainable deep neural network for identifying cardiac abnormalities using class activation map | |
| Gliner et al. | Identification of features for machine learning analysis for automatic arrhythmogenic event classification | |
| CN108836312B (en) | Clutter rejection method and system based on artificial intelligence | |
| CN117547283A (en) | Electrocardiosignal data classification method, device, computer equipment and 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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181225 |
|
| RJ01 | Rejection of invention patent application after publication |