CN108784678A - A kind of analysis method, server and the system of ectocardia beating - Google Patents
A kind of analysis method, server and the system of ectocardia beating Download PDFInfo
- Publication number
- CN108784678A CN108784678A CN201611099650.1A CN201611099650A CN108784678A CN 108784678 A CN108784678 A CN 108784678A CN 201611099650 A CN201611099650 A CN 201611099650A CN 108784678 A CN108784678 A CN 108784678A
- Authority
- CN
- China
- Prior art keywords
- qrs
- electrocardiosignal
- point
- beat
- user
- 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
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]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/318—Heart-related electrical modalities, e.g. electrocardiography [ECG]
- A61B5/346—Analysis of electrocardiograms
- A61B5/349—Detecting specific parameters of the electrocardiograph cycle
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
- A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7271—Specific aspects of physiological measurement analysis
- A61B5/7282—Event detection, e.g. detecting unique waveforms indicative of a medical condition
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Animal Behavior & Ethology (AREA)
- Veterinary Medicine (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Pathology (AREA)
- Cardiology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physiology (AREA)
- Psychiatry (AREA)
- Signal Processing (AREA)
- Evolutionary Computation (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The invention belongs to signal processing technology fields, provide a kind of analysis method, mobile terminal and the system of ectocardia beating, the method includes：Receive the electrocardiosignal for the pretreated user to be measured that mobile terminal is sent；According to the pretreated electrocardiosignal, characteristic point is obtained；According to the characteristic point, characteristic value is obtained；The arrhythmia cordis database of MIT-BIH is trained with the method for machine learning, obtains disaggregated model；With the disaggregated model to by eigenvalue cluster at feature vector classify, institute's measuring heart-beat beat cycles are judged respectively whether in ventricular ectopic beat state or whether in atrial ectopic beat state, and then judge whether user is in ventricular ectopic beat state or atrial ectopic beat state；And analysis result is issued into the mobile terminal, show analysis result in mobile terminal.The accuracy of analysis of cardiac ectopic beat and the portability of such terminal device and wearable property can be improved through the invention.
Description
Technical field
The invention belongs to signal processing technology field more particularly to a kind of analysis method of ectocardia beating, servers
And system.
Background technology
Rhythm abnormality presses its occurring principle, can divide into impulsion and form abnormal and impulse conduction exception two major classes.Dystopy
Beating is exactly that impulsion forms a kind of abnormal typical cause.The normal beats of human heart rely primarily on the self-discipline of cardiac muscle cell
Property, self-disciplining refer to there are a spontaneous depolarization and generating new round action potential in 4 phases of Single Cardiac Cell
Process.The main self-regulatory organization of human body is sinoatrial node, and other self-regulatory organizations are referred to as potential pacemaker.Under normal circumstances, sinus
Room knot can drive oppressive two ways to ensure the normal pacemaker foci status of oneself by speedily carrying out rescue work to capture and exceed the speed limit.Potential pacemaker can
To ensure that heart can be beaten with lower frequency when dysfunction occurs for sinoatrial node, ensure that the circulatory system is unlikely to paralyse, but
It is also one of risk factor simultaneously, when its self-disciplining is more than sinoatrial node, i.e.,：Arrhythmia cordis will occur in ectopic beat.
So the detection of ectocardia beating has important role, ectocardia to beat auxiliary the judgement of arrhythmia cordis
It is mainly to analyze the ecg wave form state of user's electrocardiogram to help inspection method, and current determination method accuracy rate is not high, and all
It is that large-scale hospital just has ready conditions inspection, it has not been convenient to check.
Therefore, it is necessary to propose a kind of new technical solution, to solve the above technical problems.
Invention content
In consideration of it, the embodiment of the present invention provides a kind of analysis method, server and the system of the beating of judgement ectocardia, with
Improve the accuracy of analysis of cardiac ectopic beat.
The embodiment of the present invention in a first aspect, provide a kind of analysis method of ectocardia beating, the method includes：
Receive the electrocardiosignal for the pretreated user to be measured that mobile terminal is sent；
According to the pretreated electrocardiosignal, detection obtains the characteristic point of the electrocardiosignal of the user to be measured；
The characteristic value of electrocardiosignal, composition characteristic vector are obtained by the characteristic point；
By the way that the arrhythmia cordis database of MIT-BIH is trained as training sample, disaggregated model is obtained；
Classified to described eigenvector by the disaggregated model, analyzes and determines the cardiac electrical cycle in the period
With the presence or absence of ectopic beat；
The second aspect of the embodiment of the present invention, provides a kind of server, and the server includes：
Receiving module, the electrocardiosignal of the pretreated user to be measured for receiving mobile terminal transmission；
The third aspect of the embodiment of the present invention provides a kind of analysis system of ectocardia beating, the system comprises：
Electrocardiosignal collection box, mobile terminal and server；
The electrocardiosignal collection box, the electrocardiosignal for acquiring user to be measured, and the electrocardiosignal is sent to
The mobile terminal；
The mobile terminal, for being pre-processed to the electrocardiosignal received, and by pretreated electrocardio
Signal is sent to the server；
The server, the pretreated electrocardiosignal sent for receiving the mobile terminal；According to described
Pretreated electrocardiosignal obtains characteristic point such as QRS wave starting point, terminal and the peak point of the electrocardiosignal, T crest value points
With P crest value points；According to the characteristic point, that is, QRS wave starting point, peak point and terminal, the characteristic value of the electrocardiosignal is obtained such as
The width and gradient of QRS wave；According to the characteristic point, that is, QRS wave peak point, T crest values point and P crest value points obtain the heart
Phase between phase and TP between the characteristic value such as RR of electric signal；The arrhythmia cordis database of MIT-BIH is carried out with the method for machine learning
Training obtains disaggregated model；With the disaggregated model of the acquisition to being made of the width and gradient of the characteristic value, that is, QRS wave
Feature vector is classified, and analyzes and determines whether institute's measuring heart-beat beat cycles are in ventricular ectopic beat state, and is united
Count ventricular ectopic beating heartbeat number；According to the ventricular ectopic beating heartbeat number, analyze and determine whether surveyed user is in
Ventricular ectopic beat state；With the disaggregated model of the acquisition between by the phase forms between phase and TP the characteristic value, that is, RR spy
Sign vector is classified, and analyzes and determines whether institute's measuring heart-beat beat cycles are in atrial ectopic beat state, and is counted
Room ectopic beat heartbeat number；According to the room ectopic beat heartbeat number, analyze and determine whether surveyed user is in the heart
Room ectopic beat state.And analysis result is sent to the mobile terminal, so that the mobile terminal shows described point
Analyse result.
Existing advantageous effect is the embodiment of the present invention compared with prior art：The embodiment of the present invention is first by MIT-
The arrhythmia cordis database of BIH is trained, and obtains disaggregated model.Then with the disaggregated model of the acquisition to user's heart to be measured
The characteristic value of electric signal such as QRS wave width and gradient are analyzed and determined that this process joins combination using the method for machine learning
Number carries out classification analysis, improves the accuracy and sensitivity for analyzing the ectocardia beating of user to be measured.
Description of the drawings
It to describe the technical solutions in the embodiments of the present invention more clearly, below will be to embodiment or description of the prior art
Needed in attached drawing be briefly described, it should be apparent that, the accompanying drawings in the following description be only the present invention some
Embodiment for those of ordinary skill in the art without having to pay creative labor, can also be according to these
Attached drawing obtains other attached drawings.
Fig. 1 and Fig. 2 is the implementation flow chart of the analysis method for the ectocardia beating that the embodiment of the present invention one provides；
Fig. 3 and Fig. 4 is the composition schematic diagram of server provided by Embodiment 2 of the present invention；
Fig. 5 is the composition schematic diagram of the analysis system for the ectocardia beating that the embodiment of the present invention three provides.
Specific implementation mode
In order to make the purpose , technical scheme and advantage of the present invention be clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.
Embodiment one：
Fig. 1 shows the implementation process of the analysis method for the ventricular ectopic beat that the embodiment of the present invention one provides, described
Details are as follows for implementation process：
In step S101, the electrocardiosignal for the pretreated user to be measured that mobile terminal is sent is received；
In embodiments of the present invention, server receives the electrocardio letter for the pretreated user to be measured that mobile terminal is sent
Number, the pretreatment includes but not limited to baseline and is filtered, described to go Baseline Survey for removing in electrocardiosignal
Baseline drift, it is described to be filtered for filtering out the noise in electrocardiosignal.
In embodiments of the present invention, the electrocardio for the pretreated user to be measured that server can send mobile terminal is believed
It number is preserved in a manner of date or name in the database, in order to which scholar or expert etc. carry out statistical analysis, is deeply dug
Dig other hiding features of electrocardiosignal.
In step s 102, according to the pretreated electrocardiosignal, the electrocardiosignal of the user to be measured is obtained
Characteristic point；
In embodiments of the present invention, server can obtain starting point, the terminal of QRS wave by pretreated electrocardiosignal
And peak point, the starting point for obtaining QRS wave, the step of terminal and peak point, are：
Step 1 carries out length transformation, calculation formula to QRS wave：Wherein, w
For the width of length computation, i is the sample number of iteration,For the sampling interval,yk、yk-1Respectively QRS wave
The amplitude of kth and k-1 point in data；
Step 2 determines two corner positions of the transformation of length described in step 1, as QRS wave starting point QRSonAnd terminal
QRSoff；
Step 3, the maximum of points extracted between QRS wave beginning and end is peak point QRSp。
In step s 103, server can obtain two characteristic values of QRS wave by the characteristic point of the electrocardiosignal,
That is the gradient of the width of QRS wave and QRS wave, described two eigenvalue clusters are at feature vector.
Further, the width W for obtaining QRSqrsExpression formula be：Wqrs=QRSoff-QRSon, wherein QRSonFor
The starting point of QRS wave, QRSoffFor the terminal of QRS wave.The gradient k for obtaining QRS waveqrsExpression formula be：Its
Middle k1, k2The slope being equal to for amplitude at the point of 0.7 times of peak value.
In step S104, server can by the eigenvalue cluster at feature vector classify, judge institute
State whether heartbeat beat cycles are in ventricular ectopic beat.The sorting technique is the support vector machines in machine learning, tool
Body realizes that process is as follows：
Step 1：By being trained to training sample, optimal classification model is obtained, the training sample uses MIT-
The arrhythmia cordis database of BIH；
Step 2：The disaggregated model obtained with the step 1 classifies to the S103 feature vectors obtained, sentences
Whether the secondary heartbeat beat cycles break in ventricular ectopic beat state, and counts ventricular ectopic beat heartbeat number；
Step 3：Judge whether the tested user is in room property dystopy according to the ventricular ectopic beat number of the acquisition
Pulsatile status.
In step S105, the electrocardiosignal for the pretreated user to be measured that mobile terminal is sent is received；
In embodiments of the present invention, server receives the electrocardio letter for the pretreated user to be measured that mobile terminal is sent
Number, the pretreatment includes but not limited to baseline and is filtered, described to go Baseline Survey for removing in electrocardiosignal
Baseline drift, it is described to be filtered for filtering out the noise in electrocardiosignal.
In embodiments of the present invention, the electrocardio for the pretreated user to be measured that server can send mobile terminal is believed
It number is preserved in a manner of date or name in the database, in order to which scholar or expert etc. carry out statistical analysis, is deeply dug
Dig other hiding features of electrocardiosignal.
Fig. 2 shows what the embodiment of the present invention one provided to judge whether a heart beat cycle has point of atrial ectopic beat
The implementation process of analysis method, details are as follows for the implementation process：
In step s 106, according to the pretreated electrocardiosignal, the electrocardiosignal of the user to be measured is obtained
Characteristic point；
In embodiments of the present invention, server can obtain peak point, the P waves of T waves by pretreated electrocardiosignal
Peak point.The step of acquisition T crest values point and P crest value points, is as follows：
The step of acquisition T crest values point is：
Step 1：The regions QRS having been had determined in the scheme one are rejected, i.e., zero are assigned to the regions QRS；
Step 2：The peak point of T waves is extracted using the method for sliding average；
Preferably, the window width of the moving average method is 0.1 second.
The step of acquisition P crest values point is：
Step 1：Data navigate to a T waves terminal to next QRS wave starting point, i.e., search P in the data area
Crest value point；
Step 2：Multidimensional morphology derivative (multiscale mophological are carried out to data described in step 1
Derivative it) detects, the multidimensional morphology of midpoint derives transformation and is represented by：
Wherein, s is the estimated value to P wave widths, and the data point positioned in x traversal steps one, f (t) is QRS wave data；
Step 3：Local minimum positioning is carried out to the numerical value after being converted in step 2, minimum point is P crest values
Point.
In step s 107, server can be by between the characteristic point of electrocardiosignal described in step S107 acquisition characteristic value TP
Phase, and phase between characteristic value RR, described two characteristic value composition characteristics are obtained by the characteristic point of electrocardiosignal described in step S102
Vector.
Further, phase I between the acquisition TPTPExpression formula be：Itp=Pp-Pt, wherein PpFor P crest value points, PtFor T
Crest value point, PtFor negative.The expression formula of phase is between the acquisition RR：Irr=QRSp(i)-QRSp(i-1), i.e., two neighboring QRS
The distance between peak point of wave, wherein QRSp(i) it is i-th of QRS wave peak value, i is the integer more than zero.
In step S108, server can by the eigenvalue cluster at feature vector classify, judge institute
State whether heartbeat beat cycles are in atrial ectopic beat state.The sorting technique is the supporting vector in machine learning
Machine, the specific implementation process is as follows：
Step 1：By being trained to training sample, optimal classification model is obtained, the training sample uses MIT-
The arrhythmia cordis database of BIH；
Step 2：The disaggregated model obtained with the step 1 classifies to the S108 feature vectors obtained, sentences
Whether the secondary heartbeat beat cycles break in atrial ectopic beat state, and counts atrial ectopic beat heartbeat number；
Step 3：Judge whether surveyed user is new in new house according to the atrial ectopic beat heartbeat number of the acquisition
Ectopic beat state.
Embodiment two：
Fig. 3 and Fig. 4 shows the composition schematic diagram of server provided by Embodiment 2 of the present invention, for convenience of description, only
Show with the relevant part of the embodiment of the present invention, details are as follows：
Receiving module 21, the electrocardiosignal of the pretreated user to be measured for receiving mobile terminal transmission：
Module 22 is obtained, for according to the pretreated electrocardiosignal, obtaining surveyed user's electrocardiosignal
Characteristic point；
Further, the acquisition module 22 specifically includes：
First extraction unit 221, the starting point QRS for extracting QRS waveon, terminal QRSoff, length transformation is carried out to QRS waveWherein, w is the width of length computation, and i is the sample number of iteration,Between sampling
Every,Yk-1 is respectively the amplitude of kth and k-1 point in QRS wave data；Two of length transformation turn
It is the starting point QRS of QRS wave at pointon, terminal QRSoff；
Second extraction unit 222, the peak point QRS for extracting QRS wavep, the maximum value between QRS wave beginning and end
Point is peak point QRSp。
Computing module 23, the characteristic value i.e. width of QRS wave and QRS wave for obtaining user's electrocardiosignal to be measured
Gradient；
First computing unit 231, the width W for obtaining QRS waves according to the beginning and end of the QRS waveqrs=
QRSoff-QRSon, wherein QRSonFor the starting point of QRS wave, QRSoffFor the terminal of QRS wave.
Second computing unit 232, for the peak point QRS according to the QRS wavepObtain the gradient of QRS wave, QRS wave ladder
DegreeWherein, k1, k2The slope being equal to for amplitude at the point of 0.7 times of peak value.
Analyze and determine module 24, for by the eigenvalue cluster at feature vector classify, judge the heart
Jump whether beat cycles are in ventricular ectopic beat state.
Training module 241 is trained for the arrhythmia cordis database to MIT-BIH, obtains training pattern.
Statistical module 242 is analyzed, for dividing the feature value vector of the acquisition training pattern of the acquisition
Class, analyzes and determines whether the heartbeat beat cycles are in ventricular ectopic beat state, and counts ventricular ectopic beating state
Number；
Module 243 is analyzed and determined, for according to the ventricular ectopic beating state number, whether analyzing and determining surveyed user
In ventricular ectopic beat state.
Receiving module 25, the electrocardiosignal of the pretreated user to be measured for receiving mobile terminal transmission.
Module 26 is obtained, for according to the pretreated electrocardiosignal, obtaining surveyed user's electrocardiosignal
Peak point, the P crest value points of characteristic point, that is, T waves.
Further, the acquisition module 26 specifically includes：
First extraction unit 261, for pass through moving average method to the pretreated electrocardiosignal carry out processing obtain
The peak point of T waves is obtained, the window width of the moving average method is 0.1 second；
Second extraction unit 262, for deriving transformation by carrying out multidimensional morphology to pretreated electrocardiosignalWherein, s is x times to the estimated values of P wave widths
The data point positioned in step 1 is gone through, f (t) is QRS wave data.Local minimum positioning is carried out to the numerical value after transformation, it is minimum
Value point is P crest value points.
Computing module 27, phase between phase and TP between the characteristic value i.e. RR for obtaining user's electrocardiosignal to be measured.
First computing unit 271, for determining phase I between RR according to the characteristic point of user's electrocardiosignal to be measuredrr=
QRSp(i)-QRSp(i-1), i.e., the distance between the peak point of two neighboring QRS wave, wherein QRSp(i) it is i-th of QRS wave peak
Value, i are the integer more than zero；
Second computing unit 272, for being used to be determined according to the characteristic point of user's electrocardiosignal to be measured according to described
Phase I between TPtp=Pp-Pt, wherein PpFor P crest value points, PtFor T crest value points, PtFor negative.
Analyze and determine module 28, for by the eigenvalue cluster at feature vector classify, judge the heart
Jump whether beat cycles are in ventricular ectopic beat state.
Statistical module 282 is analyzed, for dividing the feature value vector of the acquisition training pattern of the acquisition
Class, analyzes and determines whether the heartbeat beat cycles are in ventricular ectopic beat state, and counts room ectopic beat state
Number；
Module 283 is analyzed and determined, for according to the room ectopic beat state number, whether analyzing and determining surveyed user
In atrial ectopic beat state.
Embodiment three：
Fig. 5 shows the composition schematic diagram of the analysis system for the ectocardia beating that the embodiment of the present invention three provides, in order to
Convenient for explanation, illustrate only with the relevant part of the embodiment of the present invention, details are as follows：
The system comprises electrocardiosignal collection box 31, mobile terminal 32 and servers 33；
The electrocardiosignal collection box 31, the electrocardiosignal for acquiring user to be measured, and the electrocardiosignal is sent
To the mobile terminal 32；
In embodiments of the present invention, the electrocardiosignal collection box 31 is portable wearable device, can be passed by electrocardio
Sensor acquires the electrocardiosignal of user to be measured in real time, and the electrocardiosignal is wirelessly sent to the mobile terminal
32, the wireless mode includes but not limited to bluetooth, WiFi etc., and the user to be measured wears for the electrocardiosignal collection box 31
Person.
The mobile terminal 32, for being pre-processed to the electrocardiosignal received, and by the pretreated heart
Electric signal is sent to the server；
In embodiments of the present invention, the mobile terminal 32 is the terminal with display function, such as mobile phone, tablet computer
Deng, the electrocardiosignal that can be sent with electrocardiosignal collection box described in real-time display 31 and preserves the electrocardiosignal, in order to
Family is used for multiple times.
In embodiments of the present invention, the pretreatment includes but not limited to baseline and is filtered, and 0.8Hz may be used
Second order butterworth high pass filter carries out Baseline Survey, described to go Baseline Survey for removing the drift of the baseline in electrocardiosignal
It moves, 30Hz quadravalence Butterworth low passes this filters may be used and be filtered, it is described to be filtered for filtering out electrocardio
Noise in signal.Pretreated electrocardiosignal can be sent to the service by the mobile terminal 32 wirelessly
Device 33, the wireless mode include but not limited to bluetooth, WiFi etc..
The server 33, the pretreated electrocardiosignal for receiving the transmission of the mobile terminal 32；According to
The pretreated electrocardiosignal obtains characteristic point such as QRS wave starting point, terminal and the peak point of the electrocardiosignal, T wave crests
Value point and P crest value points；According to the characteristic point, that is, QRS wave starting point, peak point and terminal, the feature of the electrocardiosignal is obtained
The width and gradient of value such as QRS wave；According to the characteristic point, that is, QRS wave peak point, T crest values point and P crest value points obtain institute
State between the characteristic value such as RR of electrocardiosignal the phase between phase and TP；With the method for machine learning to the arrhythmia cordis database of MIT-BIH
It is trained, obtains disaggregated model；With the disaggregated model of the acquisition to the width and gradient group by the characteristic value, that is, QRS wave
At feature vector classify, analyze and determine institute's measuring heart-beat beat cycles whether be in ventricular ectopic beat state；
With the disaggregated model of the acquisition between being classified by the feature vector that the phase forms between phase and TP the characteristic value, that is, RR, analyze
Judge whether institute's measuring heart-beat beat cycles are in atrial ectopic beat state；And analysis result is sent to the movement
Terminal 32, so that the mobile terminal 32 shows the analysis result.
Server 33 in the embodiment of the present invention is identical as the server in embodiment two, and details are referring to above-described embodiment two
Description, details are not described herein.
The technical staff in the field can be understood that, for convenience and simplicity of description, only with above-mentioned each function
The division progress of module, can be as needed and by above-mentioned function distribution by different function moulds for example, in practical application
Block is completed, i.e., the internal structure of the described server is divided into different function modules, and hardware both may be used in above-mentioned function module
Form realize, can also be realized in the form of software.In addition, the specific name of each function module is also only to facilitate phase
Mutually difference, the protection domain being not intended to limit this application.
In conclusion the embodiment of the present invention is trained by the arrhythmia cordis database to MIT-BIH first, divided
Class model.Then with the disaggregated model of the acquisition to characteristic value such as the QRS wave width and gradient of user's electrocardiosignal to be measured into
Row analyzes and determines that this process carries out classification analysis using the method for machine learning to combination parameter, improves analysis user to be measured
Ectocardia beating accuracy and sensitivity.
Those of ordinary skill in the art be further appreciated that implement the method for the above embodiments be can
It is completed with instructing relevant hardware by program, the program can be stored in a computer read/write memory medium
In, described storage medium, including ROM/RAM, disk, CD etc..
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention
All any modification, equivalent and improvement etc., should all be included in the protection scope of the present invention made by within refreshing and principle.
Claims (9)
1. a kind of analysis method of ectocardia beating, which is characterized in that the method includes：
Receive the electrocardiosignal for the pretreated user to be measured that mobile terminal is sent；
Acquisition characteristic point is marked to the pretreated electrocardiosignal, characteristic point includes starting point, terminal and the peak of QRS wave
It is worth point, the peak point of T waves, the peak point of P waves；
According to starting point, terminal and the peak point of the characteristic point, that is, QRS wave, the width and gradient of characteristic value, that is, QRS wave are obtained；
According to the mark point, that is, QRS wave peak point, T crest values point and P crest value points obtain between characteristic value such as RR between phase and TP
Phase；
The arrhythmia cordis database of MIT-BIH is trained with the method for machine learning, obtains disaggregated model；
The feature vector being made of the width and gradient of the characteristic value, that is, QRS wave is divided with the disaggregated model of the acquisition
Class, analyzes and determines whether institute's measuring heart-beat beat cycles are in ventricular ectopic beat state, counts ventricular ectopic beating shape
State number, and according to the ventricular ectopic beating heartbeat number, analyze and determine whether surveyed user is in ventricular ectopic beat
State；
With the disaggregated model of the acquisition between being classified by the feature vector that the phase forms between phase and TP the characteristic value, that is, RR,
Judge whether institute's measuring heart-beat beat cycles are in atrial ectopic beat state, count room ectopic beat heartbeat number,
And according to the room ectopic beat heartbeat number, analyze and determine whether surveyed user is in atrial ectopic beat state；
The discriminatory analysis result is sent to the mobile terminal, so that the mobile terminal shows the analysis result.
2. according to the method described in claim 1, it is characterized in that, the starting point of the extraction QRS wave, the step of terminal and peak point
Suddenly it is：
Step 1 carries out length transformation, calculation formula to QRS wave：W is length gauge
The width of calculation, i are the sample number of iteration, For the sampling interval；
Step 2 determines two corner positions of the transformation of length described in step 1, as QRS wave starting point QRSonAnd terminal
QRSoff；
Step 3, the maximum of points extracted between QRS wave beginning and end is peak point QRSp。
3. according to the method described in claim 1, it is characterized in that, the characteristic signal QRS wave for obtaining the user to be measured
The expression formula of gradient is：Wqrs=QRSoff-QRSon, wherein QRSonFor the starting point of QRS wave, QRSoffFor the terminal of QRS wave.
4. according to the method described in claim 1, it is characterized in that, the characteristic signal QRS wave for obtaining the user to be measured
The expression formula of gradient is：Wherein k1, k2The slope being equal to for amplitude at the point of 0.7 times of peak value.
5. according to the method described in claim 1, it is characterized in that, it is described extraction T crest value points method be moving average method,
The regions QRS having had determined are rejected with the method for assigning zero first, then extract the peak point of T waves with 0.1 second window width.
6. according to the method described in claim 1, it is characterized in that, described divide two characteristic signals as feature vector
Class, sorting technique are the support vector machines in machine learning.
7. according to the method described in claim 1, it is characterized in that, described according to ectopic beat heartbeat number, analytical judgment institute
Survey user whether in ectopic beat state include：
When the ventricular ectopic beating heartbeat number meets certain preset condition, determine that the user to be measured is in ventricular different
Position pulsatile status；
When the room ectopic beat heartbeat number meets certain preset condition, determine that the user to be measured is in atrial different
Position pulsatile status.
8. a kind of server, which is characterized in that the server includes ventricular ectopic beat analysis module and atrial dystopy
Beating analysis module.
Ventricular ectopic beat analysis module is as follows：
Receiving module, the electrocardiosignal of the pretreated user to be measured for receiving mobile terminal transmission；
Module is obtained, for according to the pretreated electrocardiosignal, obtaining the characteristic point of surveyed user's electrocardiosignal；
Further, the acquisition module specifically includes：
First extraction unit, the starting point QRS for extracting QRS waveon, terminal QRSoff, length transformation is carried out to QRS waveWherein, w is the width of length computation, and i is the sample number of iteration, It is the starting point QRS of QRS wave at two inflection points converted for sampling interval, lengthb, terminal QRSe；
Second extraction unit, the peak point QRS for extracting QRS wavep, the maximum of points between QRS wave beginning and end is peak
It is worth point QRSp。
Computing module, the gradient of the characteristic value i.e. width of QRS wave and QRS wave for obtaining user's electrocardiosignal to be measured；
First computing unit, the width W for obtaining QRS wave according to the beginning and end of the QRS waveqrs=QRSoff-QRSon,
Wherein QRSonFor the starting point of QRS wave, QRSoffFor the terminal of QRS wave；
Second computing unit, for the peak point QRS according to the QRS wavepThe gradient of QRS wave is obtained, QRS wave ladder analyzes and determines
Module, for by the eigenvalue cluster at feature vector classify, judge whether the heartbeat beat cycles are in
Ventricular ectopic beat state counts the ventricular ectopic beating heartbeat number, and according to the ventricular ectopic beating heartbeat
Number, analyzes and determines whether surveyed user is in ventricular ectopic beat state.
Training module is trained for the arrhythmia cordis database to MIT-BIH, obtains training pattern；
It analyzes and determines module, for the training pattern of the acquisition to classify to the feature value vector of the acquisition, analyzes
Judge whether the heartbeat beat cycles are in ventricular ectopic beat state, count the room ectopic beat heartbeat number,
And according to the room ectopic beat heartbeat number, analyze and determine whether surveyed user is in atrial ectopic beat state；
Atrial ectopic beat is analyzed as follows：
Receiving module, the electrocardiosignal of the pretreated user to be measured for receiving mobile terminal transmission；
Module is obtained, for according to the pretreated electrocardiosignal, obtaining the characteristic point of surveyed user's electrocardiosignal
That is the peak point of T waves, P crest value points.
Further, the acquisition module specifically includes：
First extraction unit obtains T waves for carrying out processing to the pretreated electrocardiosignal by moving average method
The window width of peak point, the moving average method is 0.1 second；
Second extraction unit, for deriving transformation by carrying out multidimensional morphology to pretreated electrocardiosignalWherein, s is x times to the estimated values of P wave widths
The data point positioned in step 1 is gone through, local minimum positioning is carried out to the numerical value after transformation, minimum point is P crest values
Point.
Computing module, phase between phase and TP between the characteristic value i.e. RR for obtaining user's electrocardiosignal to be measured；
First computing unit, for determining phase I between RR according to the characteristic point of user's electrocardiosignal to be measuredrr=QRSp(i)-
QRSp(i-1), i.e., the distance between the peak point of two neighboring QRS wave, wherein QRSp(i) it is i-th of QRS wave peak value, i is big
In zero integer；
Second computing unit, for according to described for determining phase I between TP according to the characteristic point of user's electrocardiosignal to be measuredtp
=Pp-Pt, wherein PpFor P crest value points, PtFor T crest value points, PtFor negative.
Analyze and determine module, for by the eigenvalue cluster at feature vector classify, judge heartbeat beating
Whether the period is in ventricular ectopic beat state, counts ventricular ectopic beating state number, and fight according to the room property dystopy
Aroused in interest jump number, analyzes and determines whether surveyed user is in ventricular ectopic beat state；
Training module is trained for the arrhythmia cordis database to MIT-BIH, obtains training pattern；
It analyzes and determines module, for the training pattern of the acquisition to classify to the feature value vector of the acquisition, analyzes
Judge whether the heartbeat beat cycles are in atrial ectopic beat state, counts ventricular ectopic beating heartbeat number, and root
According to the ventricular ectopic beating heartbeat number, analyze and determine whether surveyed user is in ventricular ectopic beat state.
9. a kind of analysis system of ectocardia beating, which is characterized in that the system comprises：
Electrocardiosignal collection box, mobile terminal and server；
The electrocardiosignal collection box, the electrocardiosignal for acquiring user to be measured, and the electrocardiosignal is sent to described
Mobile terminal；
The mobile terminal, for being pre-processed to the electrocardiosignal received, and by pretreated electrocardiosignal
It is sent to the server；
The server, the pretreated electrocardiosignal sent for receiving the mobile terminal；According to the pre- place
Electrocardiosignal after reason obtains characteristic point such as QRS wave starting point, terminal and the peak point of the electrocardiosignal, T crest values point and P
Crest value point；According to the characteristic point, that is, QRS wave starting point, peak point and terminal, the characteristic value such as QRS of the electrocardiosignal is obtained
The width and gradient of wave；According to the characteristic point, that is, QRS wave peak point, T crest values point and P crest value points obtain the electrocardio
Phase between phase and TP between the characteristic value such as RR of signal；The arrhythmia cordis database of MIT-BIH is instructed with the method for machine learning
Practice, obtains disaggregated model；With the disaggregated model of the acquisition to the spy that is made of the width and gradient of the characteristic value, that is, QRS wave
Sign vector is classified, and analyzes and determines whether institute's measuring heart-beat beat cycles are in ventricular ectopic beat state, and is counted
Ventricular ectopic beating state number；According to the ventricular ectopic beating state number, analyze and determine whether surveyed user is in the heart
Ventricular ectopic beating state；With the disaggregated model of the acquisition between by the phase forms between phase and TP the characteristic value, that is, RR feature
Vector is classified, and analyzes and determines whether institute's measuring heart-beat beat cycles are in atrial ectopic beat state, and counts room
Property ectopic beat state number；According to the ventricular ectopic beating state number, analyze and determine whether surveyed user is in atrium
Property ectopic beat state；And analysis result is sent to the mobile terminal, so that the mobile terminal shows the analysis
As a result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611099650.1A CN108784678A (en) | 2017-05-02 | 2017-05-02 | A kind of analysis method, server and the system of ectocardia beating |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611099650.1A CN108784678A (en) | 2017-05-02 | 2017-05-02 | A kind of analysis method, server and the system of ectocardia beating |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108784678A true CN108784678A (en) | 2018-11-13 |
Family
ID=64068941
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611099650.1A Pending CN108784678A (en) | 2017-05-02 | 2017-05-02 | A kind of analysis method, server and the system of ectocardia beating |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108784678A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111281371A (en) * | 2018-12-10 | 2020-06-16 | 顺天乡大学校产学协力团 | Arrhythmia classification system and method and recording medium for executing the method |
| CN113876332A (en) * | 2021-10-27 | 2022-01-04 | 深圳大学 | Electrocardiosignal monitoring device and method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| US20150223759A1 (en) * | 2010-03-15 | 2015-08-13 | Singapore Health Services Pte Ltd. | Predicting acute cardiopulmonary events and survivability of a patient |
| US20160120430A1 (en) * | 2014-10-29 | 2016-05-05 | Khalifa University of Science, Technology, and Research | Medical device for detecting a ventricular arrhythmia event |
| CN105748063A (en) * | 2016-04-25 | 2016-07-13 | 山东大学齐鲁医院 | Intelligent arrhythmia diagnosis method based on multiple-lead and convolutional neural network |
| CN105769171A (en) * | 2014-12-26 | 2016-07-20 | 中国移动通信集团公司 | Arrhythmia detection method and device |
-
2017
- 2017-05-02 CN CN201611099650.1A patent/CN108784678A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150223759A1 (en) * | 2010-03-15 | 2015-08-13 | Singapore Health Services Pte Ltd. | Predicting acute cardiopulmonary events and survivability of a patient |
| US20160120430A1 (en) * | 2014-10-29 | 2016-05-05 | Khalifa University of Science, Technology, and Research | Medical device for detecting a ventricular arrhythmia event |
| CN105769171A (en) * | 2014-12-26 | 2016-07-20 | 中国移动通信集团公司 | Arrhythmia detection method and device |
| 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 |
| CN105748063A (en) * | 2016-04-25 | 2016-07-13 | 山东大学齐鲁医院 | Intelligent arrhythmia diagnosis method based on multiple-lead and convolutional neural network |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111281371A (en) * | 2018-12-10 | 2020-06-16 | 顺天乡大学校产学协力团 | Arrhythmia classification system and method and recording medium for executing the method |
| CN113876332A (en) * | 2021-10-27 | 2022-01-04 | 深圳大学 | Electrocardiosignal monitoring device and method |
| CN113876332B (en) * | 2021-10-27 | 2023-12-08 | 深圳大学 | Electrocardiosignal monitoring device and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11234629B2 (en) | Method and device for self-learning dynamic electrocardiography analysis employing artificial intelligence | |
| CN106214145B (en) | Electrocardiogram classification method based on deep learning algorithm | |
| AL-Ziarjawey et al. | Heart rate monitoring and PQRST detection based on graphical user interface with Matlab | |
| CN106073755A (en) | The implementation method that in a kind of miniature holter devices, atrial fibrillation identifies automatically | |
| CN108030488A (en) | The detecting system of arrhythmia cordis based on convolutional neural networks | |
| CN105550659A (en) | Real-time electrocardiogramclassification method based on random projection | |
| Wu et al. | A novel method to detect multiple arrhythmias based on time-frequency analysis and convolutional neural networks | |
| CN109691994A (en) | A kind of rhythm of the heart analysis method based on electrocardiogram | |
| CN106066933A (en) | The method obtaining atrial fibrillation identification artificial neural network weighted value matrix | |
| CN108968941A (en) | A kind of arrhythmia detection method, apparatus and terminal | |
| Šprager et al. | Feasibility of biometric authentication using wearable ECG body sensor based on higher-order statistics | |
| CN109044348A (en) | atrial fibrillation detection device and storage medium | |
| CN109288515A (en) | Periodical monitoring method and device based on premature beat signal in wearable ECG signal | |
| CN109222964A (en) | atrial fibrillation detection device and storage medium | |
| Hadiyoso et al. | Arrhythmia detection based on ECG signal using Android mobile for athlete and patient | |
| Ranjan et al. | A unified approach of ECG signal analysis | |
| CN114886388A (en) | Evaluation method and device for quality of electroencephalogram signal in anesthesia depth monitoring process | |
| CN113288168A (en) | Wearable fatigue monitoring of intelligence and early warning system | |
| CN108784678A (en) | A kind of analysis method, server and the system of ectocardia beating | |
| Venton et al. | Signal to image to classification: Transfer learning for ecg | |
| KR101510522B1 (en) | System for classification electrocardiogram signals | |
| Reddy et al. | Ecg signal characterization and correlation to heart abnormalities | |
| Bohan et al. | Ecg signal processing and human state detection based on wearable electrodes | |
| CN102058407B (en) | Method and apparatus for predicting ventricular fibrillation | |
| Xue et al. | An ECG arrhythmia classification and heart rate variability analysis system based on android platform |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20181113 |