CN107358196A - A kind of sorting technique of heart beat type, device and electrocardiogram equipment - Google Patents
A kind of sorting technique of heart beat type, device and electrocardiogram equipment Download PDFInfo
- Publication number
- CN107358196A CN107358196A CN201710564424.4A CN201710564424A CN107358196A CN 107358196 A CN107358196 A CN 107358196A CN 201710564424 A CN201710564424 A CN 201710564424A CN 107358196 A CN107358196 A CN 107358196A
- Authority
- CN
- China
- Prior art keywords
- electrocardiosignal
- heartbeat
- positioning
- waveform
- template
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
- G06F2218/10—Feature extraction by analysing the shape of a waveform, e.g. extracting parameters relating to peaks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The invention discloses a kind of sorting technique of heart beat type, device and electrocardiogram equipment, belong to medical instruments field.This method includes：The matching degree characteristic parameter of heartbeat and template heartbeat is extracted while heartbeat waveform feature parameter is extracted from electrocardiosignal again, is analyzed by decision tree classifier, so as to obtain measuring the type analysis result of heartbeat.The characteristic vector of the inventive method extraction can more show the difference of electrocardiosignal, compensate for the deficiency of single use Waveform Analysis Method or template analysis method, present invention additionally comprises the electrocardiogram equipment with this method.Present invention extraction eigenvector algorithm is simply easily achieved, and arithmetic speed is fast, and classification accuracy is high, suitable for wearable ECG real-time monitor automatic analysis diagnosis system.
Description
Technical field
The present invention relates to technical field of medical instruments, more particularly to a kind of heart beat type sorting technique, device and electrocardiogram equipment.
Background technology
In worldwide, angiocardiopathy is still main causes of death.Human body electrocardio figure signal record is
Heart cell depolarization and process of repolarization, reflect generation and the conductive process of heart electrical excitation, to a certain extent, can be objective
Reflect the physiological status at each position of heart, the evaluation for the diagnosis of heart disease, cardiac function provides important evidence.In the modern times
In medical science, the analysis and diagnosis of the automatic detection of electrocardiosignal (ECG) and analysis for angiocardiopathy have highly important
Meaning, even more numerous core components with the cardiac monitoring equipment for automatically analyzing diagnostic function.Therefore, based on electrocardiogram
Automatic detection and analysis method obtained extensive research.
All there is certain otherness in the sorting technique of the heart beat type currently existed, the spy in good grounds electrocardiogram
Waveform is levied to be analyzed, the signature waveform parameter mainly chosen has：QRS complex width, phase, ripple between waveforms amplitude, waveform
Shape form direction etc., these signature waveform parameters and the judgment threshold of fixation set in advance contrast, according to the contrast with threshold value
As a result corresponding heart beat type is judged, such as：Sinus property heartbeat, VPB, room heartbeat etc., but such method chooses fixation
Threshold value as basis for estimation, the selection of threshold value to clinical experience and analyze experience dependence it is larger, and universality is insufficient,
The effectively selection of threshold value is relatively difficult, and the unconspicuous heart beat type of difference can not effectively be differentiated.
The content of the invention
The purpose of the present invention is to overcome the shortcomings that above-mentioned, and on the one hand the present invention provides a kind of sorting technique of heart beat type,
The accuracy of classification can be improved, is advantageous to the judgement of heart beat type, for the diagnosis basis of clinical offer science, second aspect sheet
Invention provides a kind of device of heart beat type analysis method, the device have carry the device of above-mentioned heart beat type analysis method with
And memory and processor, effective carrier is provided for this method, can be more accurately to have to heart beat type
The analysis of effect, the third aspect is the invention provides a kind of electrocardiogram equipment, a kind of electrocardiogram equipment with above-mentioned heart beat type sorting technique,
Enable the patient to detect and detect at any time the electrocardiosignal of oneself using the electrocardiogram equipment, electrocardio letter can be monitored at any time by being sitting in family
Number change, such as find that electrocardiosignal can go hospitalize in time extremely, serve prevention heart infarction, coronary heart disease, heart function decline
The effect that disease is burst such as exhaust.
The embodiments of the invention provide a kind of sorting technique of heart beat type, device and electrocardiogram equipment.For the reality to disclosure
Applying some aspects of example has a basic understanding, shown below is simple summary.The summarized section is not extensive overview,
It is not intended to identify key/critical component or describes the protection domain of these embodiments.Its sole purpose is with simple shape
Some concepts are presented in formula, in this, as the preamble of following detailed description.
According to a first aspect of the present invention, there is provided a kind of sorting technique of heart beat type, comprise the following steps：
The electrocardiosignal of input is pre-processed to obtain filtering electrocardiosignal, the electrocardiosignal includes known type
Electrocardiosignal and electrocardiosignal to be detected；
Heartbeat is carried out to filtering electrocardiosignal to position to obtain positioning electrocardiosignal, the ripple of the heartbeat of extraction positioning electrocardiosignal
Shape characteristic parameter；
The positioning electrocardiosignal is normalized, obtains normalizing electrocardiosignal, extraction normalizing electrocardiosignal
Heartbeat data；
The heartbeat data of the normalizing electrocardiosignal are matched with template heartbeat data, it is special to extract the matching degree of the two
Levy parameter；
Using extraction positioning electrocardiosignal heartbeat waveform feature parameter and matching degree characteristic parameter as heartbeat spy
Sign vector is input in decision tree classifier, trains decision tree classifier, the decision tree classifier after being trained；
By the waveform feature parameter of the heartbeat of normalizing electrocardiosignal corresponding to electrocardiosignal to be detected and corresponding matching
Degree characteristic parameter is input in the decision tree classifier after the training as the characteristic vector of heartbeat, exports the heart beat type
Classification results.
By the above method, matching degree characteristic parameter is especially the increase in so that the sorting technique is more accurate.
Preferably, the electrocardiosignal of the known type includes：The electrocardiosignal of normal sinus heartbeat, supraventricular premature beat or
Electrocardiosignal, electrocardiosignal, the unassorted heartbeat electrocardiosignal of VPB of exception.
Preferably, the electrocardiosignal of input is pre-processed to obtain filtering electrocardiosignal, the pretreatment is to pass through filter
The mode of ripple is pre-processed, and the myoelectricity interference in electrocardiosignal is removed by LPF mode；Pass through high-pass filtering mode
Remove baseline drift and power frequency filtering removes Hz noise.
Preferably, described pair of filtering electrocardiosignal progress heartbeat positions to obtain positioning electrocardiosignal, including by finding the heart
The mode for waveform peak of fighting, the center of waveform position is positioned, and centered on the heart beat locations of positioning, obtain positioning electrocardio letter
Number.
Preferably, the width of the waveform feature parameter including ripple of the heartbeat of the extraction positioning electrocardiosignal, R -- R interval,
Waveform morphology.
Preferably, width, R -- R interval, the waveform morphology of the extraction ripple include：
The method being combined using equipotential line and slope threshold value, according to the heart beat locations of positioning electrocardiosignal, algorithm inspection
The beginning and end of ripple is measured, the width of ripple is calculated by way of calculating the difference of terminal and starting point；
By obtaining the R -- R interval between continuous two heartbeat waveforms of positioning electrocardiosignal after calculating positioning；
Waveform morphology is the main ripple direction of specific bit electrocardiosignal heartbeat, and the main ripple of electrocardiosignal heartbeat is positioned by extracting
Direction obtains the parameter of waveform morphology.
Preferably, the choosing method of the template heartbeat includes：
The same type heart that 3 or 3 width with last continuous ripple are less than certain threshold value is chosen from positioning electrocardiosignal
Fight；
The R -- R interval value of wherein 2 heartbeats of extraction compares, and when fiducial value is less than certain threshold value, then chooses wherein one
Individual heartbeat is as template heartbeat.
Preferably, the positioning electrocardiosignal is normalized, the normalized includes：
It is front and rear in the X-axis direction on the basis of template heartbeat waveform center by the heartbeat waveform of each positioning electrocardiosignal
The numerical value of each interception equal length；
By the heartbeat waveform of each feature electrocardiosignal, by with the basis of the same current potential of template heartbeat, taking in its Y direction
Current potential after on the basis of same current potential compares numerical value；
The amplitude of each feature electrocardiosignal heartbeat waveform is adjusted to the identical order of magnitude.
Preferably, the heartbeat data of the normalizing electrocardiosignal are matched with template heartbeat data, extracts the two
Matching degree characteristic parameter, wherein, the matching degree characteristic parameter includes calculating the heartbeat data of standard cardioelectric signal and the template heart
Standard deviation, Euclidean distance, cross-correlation set occurrence between data of fighting.
Preferably, the template heartbeat is constantly to enter Mobile state to update, when the template heartbeat that early stage is selected with it is follow-up
The same type heartbeat that the width for the ripple chosen from normalizing electrocardiosignal is less than certain threshold value is made comparisons, between the R-R for comparing the two
Time value, when change between the two is more than or equal to 40%, template heartbeat is dynamically updated.
A kind of device of electrocardiosignal sorting technique of second aspect of the present invention also body, described device include：
Pretreatment unit, for being pre-processed to obtain filtering electrocardiosignal, the electrocardio letter to the electrocardiosignal of input
Number include known type electrocardiosignal and electrocardiosignal to be detected；
Feature extraction unit, position to obtain positioning electrocardiosignal, extraction positioning for carrying out filtering electrocardiosignal heartbeat
The waveform feature parameter of the heartbeat of electrocardiosignal；
Normalization unit, for the positioning electrocardiosignal to be normalized, normalizing electrocardiosignal is obtained, is extracted
The heartbeat data of normalizing electrocardiosignal；
Matching degree unit, for the heartbeat data of the normalizing electrocardiosignal to be matched with template heartbeat data, carry
Take the matching degree characteristic parameter of the two；
Training unit, for by extraction positioning electrocardiosignal heartbeat waveform feature parameter and matching degree characteristic parameter
Characteristic vector as heartbeat is input in decision tree classifier, trains decision tree classifier, the decision tree after being trained point
Class device；
Taxon, for by the waveform feature parameter of the heartbeat of normalizing electrocardiosignal corresponding to electrocardiosignal to be detected
It is input in the decision tree classifier after the training, exports as the characteristic vector of heartbeat with corresponding matching degree characteristic parameter
The classification storage medium of the classification results of the heart beat type.
The above-mentioned device make it that measurement result is more accurate.
Preferably, the pretreatment unit, for being pre-processed to obtain filtering electrocardiosignal to the electrocardiosignal of input,
The electrocardiosignal includes the electrocardiosignal of known type and electrocardiosignal to be detected, is pre-processed by way of filtering,
The myoelectricity interference in electrocardiosignal is removed by LPF mode；Baseline drift and power frequency are removed by high-pass filtering mode
Filtering removes Hz noise, and the electrocardiosignal of the known type includes：Electrocardiosignal, the supraventricular premature beat of normal sinus heartbeat
Or electrocardiosignal, electrocardiosignal, the unassorted heartbeat electrocardiosignal of VPB of exception.
Preferably, the feature extraction unit, for carrying out heartbeat positioning to described pair of filtering electrocardiosignal, positioned
Electrocardiosignal, by way of finding heartbeat waveform peak, the center of waveform position is positioned, and using the heart beat locations of positioning in
The heart, obtain positioning electrocardiosignal, the waveform feature parameter of the heartbeat of extraction positioning electrocardiosignal, waveform feature parameter includes ripple
Width, R -- R interval, waveform morphology, including：
The method being combined using equipotential line and slope threshold value, according to the heart beat locations of positioning electrocardiosignal, algorithm inspection
The beginning and end of ripple is measured, the width of ripple is calculated by way of calculating the difference of terminal and starting point；
By obtaining the R -- R interval between continuous two heartbeat waveforms of positioning electrocardiosignal after calculating positioning；
Waveform morphology is the main ripple direction of specific bit electrocardiosignal heartbeat, and the main ripple of electrocardiosignal heartbeat is positioned by extracting
Direction obtains the parameter of waveform morphology.
Preferably, it is characterised in that also including template heartbeat unit, for choosing template heartbeat, including：
The same type heart that 3 or 3 width with last continuous ripple are less than certain threshold value is chosen from positioning electrocardiosignal
Fight；
The R -- R interval value of wherein 2 heartbeats of extraction compares, and when fiducial value is less than certain threshold value, then chooses wherein one
Individual heartbeat is as template heartbeat.
Preferably, normalization unit, for the positioning electrocardiosignal to be normalized, normalizing electrocardio letter is obtained
Number, the heartbeat data of extraction normalizing electrocardiosignal；The normalized includes：
It is front and rear in the X-axis direction on the basis of template heartbeat waveform center by the heartbeat waveform of each positioning electrocardiosignal
The numerical value of each interception equal length；
By the heartbeat waveform of each feature electrocardiosignal, by with the basis of the same current potential of template heartbeat, taking in its Y direction
Current potential after on the basis of same current potential compares numerical value；
The amplitude of each feature electrocardiosignal heartbeat waveform is adjusted to the identical order of magnitude.
Preferably, matching degree unit, for the heartbeat data of the normalizing electrocardiosignal and template heartbeat data to be carried out
Matching, extracts the matching degree characteristic parameter of the two；The matching degree characteristic parameter includes the heart rate for calculating standard cardioelectric signal
According to the standard deviation between template heartbeat data, Euclidean distance, cross-correlation set occurrence.
A kind of electrocardiogram equipment of third aspect present invention also body, including heart beat type sorting technique described above, device
Electrocardiogram equipment.
Technical scheme provided in an embodiment of the present invention can include the following benefits：
Present invention incorporates the wave character of electrocardiosignal and template matches degree feature, the combination of these features can be more
More fully reflect the otherness of different heart beat type electrocardiosignals comprehensively, so as to more effectively detect different heart beat types
Electrocardiosignal, while sinus property heartbeat, supraventricular heartbeat and room property heartbeat are classified using decision tree classifier, generated
Decision tree classifier model it is simple in rule, should be readily appreciated that and realize, improve the speed and accuracy of classification.
It should be appreciated that the general description and following detailed description of the above are only exemplary and explanatory, not
Can the limitation present invention.
Brief description of the drawings
Accompanying drawing herein is merged in specification and forms the part of this specification, shows the implementation for meeting the present invention
Example, and for explaining principle of the invention together with specification.
Fig. 1 is a kind of schematic flow sheet one of the sorting technique of heart beat type according to an exemplary embodiment；
Fig. 2 is a kind of schematic flow sheet two of the sorting technique of heart beat type according to an exemplary embodiment；
Fig. 3 is the schematic device of the electrocardiosignal sorting technique according to an exemplary embodiment；
Fig. 4 is the decision tree classifier in a kind of sorting technique of heart beat type according to an exemplary embodiment
The flow chart of sorting technique.
Embodiment
The following description and drawings fully show specific embodiments of the present invention, to enable those skilled in the art to
Put into practice them.Embodiment only represents possible change.Unless explicitly requested, otherwise single components and functionality is optional, and
And the order of operation can change.The part of some embodiments and feature can be included in or replace other embodiments
Part and feature.The scope of embodiment of the present invention includes the gamut of claims, and the institute of claims
There is obtainable equivalent.Herein, each embodiment can individually or generally be represented that this is only with term " invention "
It is merely for convenience, and if in fact disclosing the invention more than one, it is not meant to automatically limit the scope of the application
For any single invention or inventive concept.Herein, such as first and second or the like relational terms are used only for one
Entity or operation make a distinction with another entity or operation, exist without requiring or implying between these entities or operation
Any actual relation or order.Moreover, term " comprising ", "comprising" or its any other variant be intended to it is non-exclusive
Property includes, so that process, method or equipment including a series of elements not only include those key elements, but also including
The other element being not expressly set out.Each embodiment herein is described by the way of progressive, and each embodiment stresses
Be all difference with other embodiment, between each embodiment identical similar portion mutually referring to.For implementing
For structure, product etc. disclosed in example, due to its with embodiment disclosed in part it is corresponding, so fairly simple, the phase of description
Part is closed referring to method part illustration.
Flow chart as shown in Figure 1, the invention discloses a kind of sorting technique of heart beat type, this method includes following step
Suddenly：
S101：The electrocardiosignal of input is pre-processed, the electrocardiosignal of the electrocardiosignal including known type and
Electrocardiosignal to be detected；
S102：Heartbeat is carried out to pretreated electrocardiosignal to position to obtain feature electrocardiosignal, extraction feature electrocardio letter
Number heartbeat waveform feature parameter；
S103：The feature electrocardiosignal is normalized, obtains standard cardioelectric signal, extraction standard electrocardio letter
Number heartbeat；
S104：The heartbeat of the standard cardioelectric signal is matched with template heartbeat, extracts the matching degree feature of the two
Parameter；
S105：Using the waveform feature parameter and matching degree characteristic parameter of the heartbeat of the feature electrocardiosignal of extraction as heartbeat
Characteristic vector be input in decision tree classifier, train decision tree classifier, the decision tree classifier after being trained；
S106：By the waveform feature parameter of the heartbeat of feature electrocardiosignal corresponding to electrocardiosignal to be detected and corresponding
Matching degree characteristic parameter is input in the decision tree classifier after the training as the characteristic vector of heartbeat, exports the heartbeat class
The classification results of type.
The present invention realizes is divided into 4 classes according to cardiac electrical excitement origin is different heartbeat：Normal sinus heartbeat (Normal
Sinus rhythm, N or NSR), supraventricular premature beat or abnormal (Premature or ectopic supraventricular
Beat, S), VPB (Prematureventricular contraction, V), unfiled heartbeat
(Unclassifiable beat, Q), i.e., heart beat type is divided into four kinds of N-type, S types, V-type, Q types different types, in above-mentioned
The electrocardiosignal of the known type of input is different types of electrocardiosignal in above-mentioned 4.
As shown in Fig. 2 according to the sorting technique of above-mentioned heart beat type, further, the known electrocardio of the input is believed
Number carrying out pretreatment is pre-processed by way of filtering, and the electrocardiosignal of described pair of input is filtered processing, step
S101 specifically also includes：
S1011：Myoelectricity interference in electrocardiosignal is removed by LPF；
S1012：Baseline drift is removed by high-pass filtering and power frequency filtering removes Hz noise.
Heartbeat positioning is carried out to pretreated electrocardiosignal, is by way of finding heartbeat waveform peak, positions ripple
The center of shaped position, and centered on the heart beat locations of positioning, extract the data slot of QRS complex, so more can accurately
Heartbeat is positioned, the data of extraction are more accurate, so as to obtain feature electrocardiosignal.
Extract the waveform feature parameter of the heartbeat of feature electrocardiosignal, by extract the width of QRS complex, R -- R interval,
Characteristic parameter of three parameters of waveform morphology as waveform；Wave character is extracted in QRS complex data slot.
The waveform feature parameter extraction of the heartbeat of feature electrocardiosignal and its computational methods, including：
1st, the width of ripple is extracted
Because normal heartbeat and supraventricular heartbeat are respectively provided with the QRS wave form of normal conduction, and the QRS wave shape of room property heartbeat
Roomy deformity, so the width of QRS complex can be used as a judge index.
The beginning and end of QRS complex is detected by algorithm, its difference is QRS width.
The algorithm measured to the width of QRS complex includes：The side being combined using equipotential line and slope threshold value
Method, QRS starting point and initial point can be accurately positioned out.
2nd, R -- R interval is extracted
Because supraventricular premature beat or exception and VPB go out in advance mostly in time compared to normal sinus heartbeat
It is existing, so the R -- R interval between heartbeat can also be used as a judging quota.
By the heart beat locations navigated to, the R -- R interval between the continuous heartbeat of each two can be obtained.
3rd, waveform morphology is extracted
Most significantly it is exactly that main ripple has positive and negative point because ecg wave form has different forms.When one in same sequence
When the main wave morphology of individual waveform and the main wave morphology of another waveform have significantly different, then two kinds of waveforms can be judged as two kinds
Type.What the width of QRS complex embodied is the wave character of current detection heartbeat, and what R -- R interval embodied is between adjacent heartbeat
Cycle transition relation, waveform morphology embody is difference between current heart and other types heartbeat feature.
Further, in order to extract match parameter between the heartbeat of feature electrocardiosignal and template heartbeat, it is necessary to greatest extent
Other differences between the two waveform are eliminated, therefore each heartbeat sequence are normalized, at the normalization
Reason comprises the following steps：
1st, the data alignd in X-axis
The heart beat locations of next feature electrocardiosignal are oriented using each heartbeat as the alignment standard in X-axis, and it is preceding
The data of equal length are respectively intercepted afterwards.
2nd, the baseline of unified Y-axis
Due to external interference or some other factor, the equipotential line of the heartbeat of each feature electrocardiosignal is not same
On one horizontal line, this largely exacerbates the difference between waveform.In normalized Y-axis data, make each special
It is all the fiducial value on the basis of its equipotential line to levy the data in the heartbeat of electrocardiosignal.
3rd, the order of magnitude of unified QRS complex amplitude
When carrying out ecg signal acquiring, because breathing or the interference of other factors, the amplitude of waveform have larger ripple
It is dynamic, larger difference between waveform can be caused, the essence of this species diversity is not the morphological differences of ecg wave form, is unfavorable for heartbeat
Classification.When entering line amplitude and unifying, the amplitude of the heartbeat of each feature electrocardiosignal is uniformly arrived into the identical order of magnitude.
Above-mentioned normalized is carried out to the feature electrocardiosignal, so as to obtain standard cardioelectric signal, extraction standard
The heartbeat of electrocardiosignal；
In order to calculate the matching degree characteristic parameter between the heartbeat of standard cardioelectric signal and template heartbeat, template need to be first chosen
Heartbeat, the selection of template heartbeat comprise the following steps：
The heartbeat that 3 or more than 3 continuous QRS width are less than 100ms~130ms is chosen from standard cardioelectric signal,
When it is a kind of heartbeat in 4 kinds of known types to detect heart beat type, and compare 2 R -- R interval values, when both it
Between change be less than 40% when, choose one of heartbeat as template heartbeat.
Template heartbeat is not constant always, in order to improve the degree of accuracy of the classification of heart beat type, it is necessary in real time more
New template heartbeat, the update method of template heartbeat include：When above-mentioned 2 R -- R interval values of comparison, the change when between the two is more than
During equal to 40%, template heartbeat is dynamically updated.
Further, the major parameter and its computational methods of matching degree characteristic parameter, including：
1st, standard deviation
What standard deviation embodied is the dispersion degree of a data set, and what is calculated here is measurement heartbeat and template heartbeat
The dispersion degree of difference between the standard deviation of data difference, i.e. two heartbeats, thus energy dissolve the difference between the two waveforms
It is different.
By following calculation formula, template matches are carried out, calculate the standard deviation of difference between measurement heartbeat and template heartbeat.
Formula is as follows：
Wherein, xiBe test heartbeat data, yiIt is the data of template heartbeat, n is the length of heartbeat data.
2nd, Euclidean distance
Euclidean distance is most common distance metric, and measurement is absolute distance in space between each point, and distance is more remote
Illustrate that the difference between individual is bigger.The calculation formula of Euclidean distance is as follows：
Wherein, xiBe test heartbeat data, yiIt is the data of template heartbeat, n is the length of heartbeat data.
3rd, cross-correlation coefficient
Cross-correlation coefficient is the statistical indicator for weighing dependency relation level of intimate between two variables.Cross-correlation coefficient
R span is [- 1,1], r>0 represents positive correlation, r<0 represents negatively correlated, | r | illustrate the height of degree of correlation between variable
It is low.Distinguishingly, r=1 is referred to as perfect positive correlation, and r=-1 is referred to as perfect negative correlation, and r=0 is referred to as uncorrelated.Generally | r | it is more than
When 0.8, it is believed that two variables have very strong linear dependence.
Wherein, xiBe test heartbeat data, yiIt is the data of template heartbeat, n is the length of heartbeat data.
Spy using the waveform feature parameter of the heartbeat of the feature electrocardiosignal of extraction and matching degree characteristic parameter as heartbeat
Sign vector is input in decision tree classifier, trains decision tree classifier, the decision tree classifier after being trained；
Decision tree classifier as shown in Figure 4 is a tree-shaped decision diagram, and it represents characteristics of objects attribute and object class
A kind of mapping between not, each nonleaf node represent the test on a characteristic attribute, and each branch represents this feature category
Property output in some codomain, and each leaf node deposits a beat classification classification.Decision Classfication is carried out using decision tree
Process be exactly since root node, test corresponding characteristic attribute in item to be sorted, and output branch is selected according to its value, directly
To leaf node is reached, the result using the classification of leaf node storage as beat classification.
By the waveform feature parameter of the heartbeat of feature electrocardiosignal corresponding to electrocardiosignal to be detected and corresponding matching
Degree characteristic parameter is input in the decision tree classifier after the training as the characteristic vector of heartbeat, exports the heart beat type
Classification results, that is, judge it is that heartbeat to be detected be one kind in N-type, S types, V-type, the type of Q types 4.
Above-mentioned decision tree classifier is trained by multi-group data, trains the decision tree classifier of accuracy of judgement, will be treated
Thought-read electric signal is inputted in the decision tree classifier trained to this, passes through the program, you can judges it is which kind of heart beat type.
Second aspect according to embodiments of the present invention, there is provided a kind of device of electrocardiosignal sorting technique, as shown in figure 3,
Described device includes：
Pretreatment unit 201, for being pre-processed to obtain filtering electrocardiosignal, the electrocardio to the electrocardiosignal of input
Signal includes the electrocardiosignal of known type and electrocardiosignal to be detected；
Feature extraction unit 202, position to obtain positioning electrocardiosignal for carrying out filtering electrocardiosignal heartbeat, carry
Normalization unit 203, for the positioning electrocardiosignal to be normalized, normalizing electrocardiosignal is obtained,
Extract the heartbeat data of normalizing electrocardiosignal；
Matching degree unit 204, for the heartbeat data of the normalizing electrocardiosignal to be matched with template heartbeat data,
Extract the matching degree characteristic parameter of the two；
Training unit 205, for by extraction positioning electrocardiosignal heartbeat waveform feature parameter and matching degree feature
Parameter is input in decision tree classifier as the characteristic vector of heartbeat, trains decision tree classifier, the decision-making after being trained
Tree Classifier；
Taxon 206, for by the wave character of the heartbeat of normalizing electrocardiosignal corresponding to electrocardiosignal to be detected
Parameter and corresponding matching degree characteristic parameter are input in the decision tree classifier after the training as the characteristic vector of heartbeat,
Export the classification storage medium of the classification results of the heart beat type.
Further, pretreatment unit 201 as described above, is filtered for being pre-processed to the electrocardiosignal of input
Electrocardiosignal, the electrocardiosignal include the electrocardiosignal of known type and electrocardiosignal to be detected, entered by way of filtering
Row pretreatment, the myoelectricity interference in electrocardiosignal is removed by LPF mode；Baseline is removed by high-pass filtering mode to float
Move and power frequency filtering removes Hz noise, the electrocardiosignal of the known type includes：The electrocardiosignal of normal sinus heartbeat,
Supraventricular premature beat or the electrocardiosignal of exception, electrocardiosignal, the unassorted heartbeat electrocardiosignal of VPB.
Further, the feature extraction unit 202, for carrying out heartbeat positioning to described pair of filtering electrocardiosignal, obtain
Electrocardiosignal is positioned, by way of finding heartbeat waveform peak, positions the center of waveform position, and with the heart beat locations of positioning
Centered on, obtain positioning electrocardiosignal, the waveform feature parameter of the heartbeat of extraction positioning electrocardiosignal, waveform feature parameter includes
Width, R -- R interval, the waveform morphology of ripple, including：
The method being combined using equipotential line and slope threshold value, according to the heart beat locations of positioning electrocardiosignal, algorithm inspection
The beginning and end of ripple is measured, the width of ripple is calculated by way of calculating the difference of terminal and starting point；
By obtaining the R -- R interval between continuous two heartbeat waveforms of positioning electrocardiosignal after calculating positioning；
Waveform morphology is the main ripple direction of specific bit electrocardiosignal heartbeat, and the main ripple of electrocardiosignal heartbeat is positioned by extracting
Direction obtains the parameter of waveform morphology.
On the basis of more than, in addition to template heartbeat unit 203, for choosing template heartbeat, including：From positioning electrocardio letter
The same type heartbeat that 3 or 3 width with last continuous ripple are less than certain threshold value is chosen in number；
The R -- R interval value of wherein 2 heartbeats of extraction compares, and when fiducial value is less than certain threshold value, then chooses wherein one
Individual heartbeat is as template heartbeat.
Further, the normalization unit 204, for the positioning electrocardiosignal to be normalized, returned
One electrocardiosignal, extract the heartbeat data of normalizing electrocardiosignal；The normalized includes：
It is front and rear in the X-axis direction on the basis of template heartbeat waveform center by the heartbeat waveform of each positioning electrocardiosignal
The numerical value of each interception equal length；
By the heartbeat waveform of each feature electrocardiosignal, by with the basis of the same current potential of template heartbeat, taking in its Y direction
Current potential after on the basis of same current potential compares numerical value；
The amplitude of each feature electrocardiosignal heartbeat waveform is adjusted to the identical order of magnitude.
Further, the matching degree unit 205, for by the heartbeat data of the normalizing electrocardiosignal and template heart rate
According to being matched, the matching degree characteristic parameter of the two is extracted；The matching degree characteristic parameter includes calculating standard cardioelectric signal
Standard deviation, Euclidean distance, cross-correlation set occurrence between heartbeat data and template heartbeat data.
The third aspect according to embodiments of the present invention, there is provided a kind of electrocardiogram equipment, the electrocardiogram equipment include above-mentioned heart beat type
Sorting technique.
Further, the electrocardiogram equipment also includes the device of above-mentioned electrocardiosignal sorting technique on this basis.
It should be appreciated that the invention is not limited in the flow and structure for being described above and being shown in the drawings,
And various modifications and changes can be being carried out without departing from the scope.The scope of the present invention is only limited by appended claim
System.
Claims (17)
1. a kind of sorting technique of heart beat type, it is characterised in that comprise the following steps：
The electrocardiosignal of input is pre-processed to obtain filtering electrocardiosignal, the electrocardiosignal includes the electrocardio of known type
Signal, electrocardiosignal to be detected；
Carry out heartbeat to filtering electrocardiosignal to position to obtain positioning electrocardiosignal, the waveform of the heartbeat of extraction positioning electrocardiosignal is special
Levy parameter；
The positioning electrocardiosignal is normalized, normalizing electrocardiosignal is obtained, extracts the heartbeat of normalizing electrocardiosignal
Data；
The heartbeat data of the normalizing electrocardiosignal are matched with template heartbeat data, extract the matching degree feature ginseng of the two
Number；
Using extraction positioning electrocardiosignal heartbeat waveform feature parameter and matching degree characteristic parameter as heartbeat feature to
Amount is input in decision tree classifier, trains decision tree classifier, the decision tree classifier after being trained；
The waveform feature parameter of the heartbeat of normalizing electrocardiosignal corresponding to electrocardiosignal to be detected and corresponding matching degree is special
Sign parameter is input in the decision tree classifier after the training as the characteristic vector of heartbeat, exports the classification of the heart beat type
As a result.
2. the sorting technique of heart beat type according to claim 1, it is characterised in that the electrocardiosignal of the known type
Including：The electrocardiosignal of normal sinus heartbeat, the electrocardiosignal of supraventricular premature beat or exception, VPB electrocardiosignal, its
His unfiled heartbeat electrocardiosignal.
3. the sorting technique of heart beat type according to claim 1, it is characterised in that carried out in advance to the electrocardiosignal of input
Processing is obtained filtering electrocardiosignal, and the pretreatment is pre-processed by way of filtering, is gone by LPF mode
Except the myoelectricity interference in electrocardiosignal；Baseline drift is removed by high-pass filtering mode and power frequency filtering removes Hz noise.
4. the sorting technique of heart beat type according to claim 1, it is characterised in that described pair of filtering electrocardiosignal is carried out
Heartbeat positions to obtain positioning electrocardiosignal, including by way of finding heartbeat waveform peak, positions the center of waveform position, and
Centered on the heart beat locations of positioning, obtain positioning electrocardiosignal.
5. according to the sorting technique of any described heart beat type of claim 1 or 4, it is characterised in that the extraction positioning heart
The waveform feature parameter of the heartbeat of electric signal includes width, R -- R interval, the waveform morphology of ripple.
6. the sorting technique of heart beat type according to claim 5, it is characterised in that between the width for extracting ripple, R-R
Phase, waveform morphology include：
The method being combined using equipotential line and slope threshold value, according to the heart beat locations of positioning electrocardiosignal, algorithm detects
The beginning and end of ripple, the width of ripple is calculated by way of calculating the difference of terminal and starting point；
By obtaining the R -- R interval between continuous two heartbeat waveforms of positioning electrocardiosignal after calculating positioning；
Waveform morphology is the main ripple direction of specific bit electrocardiosignal heartbeat, and the main ripple direction of electrocardiosignal heartbeat is positioned by extracting
Obtain the parameter of waveform morphology.
7. the sorting technique of heart beat type according to claim 1, it is characterised in that the choosing method of the template heartbeat
Including：
The same type heartbeat that 3 or 3 width with last continuous ripple are less than certain threshold value is chosen from positioning electrocardiosignal；
The R -- R interval value of wherein 2 heartbeats of extraction compares, and when fiducial value is less than certain threshold value, then chooses one of heart
Fight as template heartbeat.
8. the sorting technique of heart beat type according to claim 7, it is characterised in that carried out to the positioning electrocardiosignal
Normalized, the normalized include：
By the heartbeat waveform of each positioning electrocardiosignal, on the basis of template heartbeat waveform center, front and rear each section in the X-axis direction
Take the numerical value of equal length；
By the heartbeat waveform of each feature electrocardiosignal, by with the basis of the same current potential of template heartbeat, taking in its Y direction with same
Current potential after on the basis of one current potential compares numerical value；
The amplitude of each feature electrocardiosignal heartbeat waveform is adjusted to the identical order of magnitude.
9. the sorting technique of heart beat type according to claim 1, it is characterised in that by the heart of the normalizing electrocardiosignal
Data of fighting are matched with template heartbeat data, extract the matching degree characteristic parameter of the two, wherein, the matching degree characteristic parameter
Standard deviation, Euclidean distance, cross-correlation set occurrence between heartbeat data and template heartbeat data including calculating standard cardioelectric signal.
10. the sorting technique of heart beat type according to claim 7, it is characterised in that the template heartbeat is constantly to enter
Mobile state renewal, when the width of the template heartbeat that early stage is selected and the follow-up ripple chosen from normalizing electrocardiosignal is less than one
The same type heartbeat for determining threshold value is made comparisons, and compares the R -- R interval value of the two, when change between the two is more than or equal to 40%,
Template heartbeat is dynamically updated.
11. a kind of device of electrocardiosignal sorting technique, it is characterised in that described device includes：
Pretreatment unit, for being pre-processed to obtain filtering electrocardiosignal, the electrocardiosignal bag to the electrocardiosignal of input
Include the electrocardiosignal of known type and electrocardiosignal to be detected；
Feature extraction unit, position to obtain positioning electrocardiosignal, extraction positioning electrocardio for carrying out filtering electrocardiosignal heartbeat
The waveform feature parameter of the heartbeat of signal；
Normalization unit, for the positioning electrocardiosignal to be normalized, normalizing electrocardiosignal is obtained, extracts normalizing
The heartbeat data of electrocardiosignal；
Matching degree unit, for the heartbeat data of the normalizing electrocardiosignal to be matched with template heartbeat data, extraction two
The matching degree characteristic parameter of person；
Training unit, for using extraction positioning electrocardiosignal heartbeat waveform feature parameter and matching degree characteristic parameter as
The characteristic vector of heartbeat is input in decision tree classifier, trains decision tree classifier, the decision tree classifier after being trained；
Taxon, for by the waveform feature parameter of the heartbeat of normalizing electrocardiosignal corresponding to electrocardiosignal to be detected and right
The matching degree characteristic parameter answered is input in the decision tree classifier after the training as the characteristic vector of heartbeat, exports the heart
The classification storage medium of the classification results for type of fighting.
12. the device of electrocardiosignal sorting technique according to claim 11, it is characterised in that the pretreatment unit,
For being pre-processed to obtain filtering electrocardiosignal to the electrocardiosignal of input, the electrocardiosignal includes the electrocardio of known type
Signal and electrocardiosignal to be detected, are pre-processed by way of filtering, are removed by LPF mode in electrocardiosignal
Myoelectricity interference；Baseline drift is removed by high-pass filtering mode and power frequency filtering removes Hz noise, the known type
Electrocardiosignal include：The electrocardiosignal of normal sinus heartbeat, the electrocardiosignal of supraventricular premature beat or exception, the heart of VPB
Electric signal, unassorted heartbeat electrocardiosignal.
13. the device of electrocardiosignal sorting technique according to claim 11, it is characterised in that the feature extraction list
Member, for carrying out heartbeat positioning to described pair of filtering electrocardiosignal, obtain positioning electrocardiosignal, by finding heartbeat waveform peak
Mode, position the center of waveform position, and centered on the heart beat locations of positioning, obtain positioning electrocardiosignal, extraction positioning
The waveform feature parameter of the heartbeat of electrocardiosignal, waveform feature parameter includes width, R -- R interval, the waveform morphology of ripple, wherein wrapping
Include：
The method being combined using equipotential line and slope threshold value, according to the heart beat locations of positioning electrocardiosignal, algorithm detects
The beginning and end of ripple, the width of ripple is calculated by way of calculating the difference of terminal and starting point；
By obtaining the R -- R interval between continuous two heartbeat waveforms of positioning electrocardiosignal after calculating positioning；
Waveform morphology is the main ripple direction of specific bit electrocardiosignal heartbeat, and the main ripple direction of electrocardiosignal heartbeat is positioned by extracting
Obtain the parameter of waveform morphology.
14. the device of electrocardiosignal sorting technique according to claim 11, it is characterised in that also including template heartbeat list
Member, for choosing template heartbeat, including：
The same type heartbeat that 3 or 3 width with last continuous ripple are less than certain threshold value is chosen from positioning electrocardiosignal；
The R -- R interval value of wherein 2 heartbeats of extraction compares, and when fiducial value is less than certain threshold value, then chooses one of heart
Fight as template heartbeat.
15. the device of electrocardiosignal sorting technique according to claim 11, it is characterised in that normalization unit, be used for
The positioning electrocardiosignal is normalized, normalizing electrocardiosignal is obtained, extracts the heartbeat data of normalizing electrocardiosignal；
The normalized includes：
By the heartbeat waveform of each positioning electrocardiosignal, on the basis of template heartbeat waveform center, front and rear each section in the X-axis direction
Take the numerical value of equal length；
By the heartbeat waveform of each feature electrocardiosignal, by with the basis of the same current potential of template heartbeat, taking in its Y direction with same
Current potential after on the basis of one current potential compares numerical value；
The amplitude of each feature electrocardiosignal heartbeat waveform is adjusted to the identical order of magnitude.
16. the device of electrocardiosignal sorting technique according to claim 11, it is characterised in that matching degree unit, be used for
The heartbeat data of the normalizing electrocardiosignal are matched with template heartbeat data, extract the matching degree characteristic parameter of the two；
Standard deviation, Europe between heartbeat data and template heartbeat data of the matching degree characteristic parameter including calculating standard cardioelectric signal
Family name's distance, cross-correlation set occurrence.
A kind of 17. electrocardiogram equipment, it is characterised in that the electrocardio including heart beat type sorting technique described in claim 1-16, device
Instrument.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710564424.4A CN107358196B (en) | 2017-07-12 | 2017-07-12 | Heart beat type classification method and device and electrocardiograph |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710564424.4A CN107358196B (en) | 2017-07-12 | 2017-07-12 | Heart beat type classification method and device and electrocardiograph |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107358196A true CN107358196A (en) | 2017-11-17 |
| CN107358196B CN107358196B (en) | 2020-11-10 |
Family
ID=60292566
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710564424.4A Expired - Fee Related CN107358196B (en) | 2017-07-12 | 2017-07-12 | Heart beat type classification method and device and electrocardiograph |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107358196B (en) |
Cited By (28)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107714023A (en) * | 2017-11-27 | 2018-02-23 | 乐普（北京）医疗器械股份有限公司 | Static ecg analysis method and apparatus based on artificial intelligence self study |
| CN107736888A (en) * | 2017-11-27 | 2018-02-27 | 中山大学 | A kind of method and system that real-time cardiac monitoring is carried out using wearable ecg equipment |
| CN107837082A (en) * | 2017-11-27 | 2018-03-27 | 乐普（北京）医疗器械股份有限公司 | Electrocardiogram automatic analysis method and device based on artificial intelligence self study |
| CN107951485A (en) * | 2017-11-27 | 2018-04-24 | 乐普（北京）医疗器械股份有限公司 | Ambulatory ECG analysis method and apparatus based on artificial intelligence self study |
| CN107981858A (en) * | 2017-11-27 | 2018-05-04 | 乐普（北京）医疗器械股份有限公司 | Electrocardiogram heartbeat automatic recognition classification method based on artificial intelligence |
| CN108491769A (en) * | 2018-03-08 | 2018-09-04 | 四川大学 | Atrial fibrillation sorting technique based on phase between RR and multiple characteristic values |
| CN108903935A (en) * | 2018-07-11 | 2018-11-30 | 上海夏先机电科技发展有限公司 | A kind of ventricular premature beat recognition methods, identifying system and electronic equipment |
| CN108937915A (en) * | 2018-07-24 | 2018-12-07 | 杭州质子科技有限公司 | The recognition methods of premature beat in a kind of dynamic ECG |
| CN109077719A (en) * | 2018-07-05 | 2018-12-25 | 广州视源电子科技股份有限公司 | signal identification method, device, equipment and storage medium |
| CN109303559A (en) * | 2018-11-01 | 2019-02-05 | 杭州质子科技有限公司 | A kind of dynamic ECG beat classification method promoting decision tree based on gradient |
| CN109303561A (en) * | 2018-11-01 | 2019-02-05 | 杭州质子科技有限公司 | It is a kind of to clap the recognition methods clapped with the abnormal heart based on the artifact heart of misclassification and supervised learning |
| CN109431488A (en) * | 2018-09-28 | 2019-03-08 | 上海优加利健康管理有限公司 | The recognition methods of typical data in a kind of electrocardiogram dynamic real-time analysis data |
| CN109691995A (en) * | 2019-01-31 | 2019-04-30 | 英菲泰克(天津)科技有限公司 | The arrhythmia cordis real-time data analysis system and method for single lead electrocardiogram |
| WO2019100561A1 (en) * | 2017-11-27 | 2019-05-31 | 乐普（北京）医疗器械股份有限公司 | Artificial intelligence-based interference recognition method for electrocardiogram |
| CN109871742A (en) * | 2018-12-29 | 2019-06-11 | 安徽心之声医疗科技有限公司 | A kind of electrocardiosignal localization method based on attention Recognition with Recurrent Neural Network |
| CN109948396A (en) * | 2017-12-20 | 2019-06-28 | 深圳市理邦精密仪器股份有限公司 | A kind of beat classification method, beat classification device and electronic equipment |
| CN111053551A (en) * | 2019-12-27 | 2020-04-24 | 深圳邦健生物医疗设备股份有限公司 | RR interval electrocardio data distribution display method, device, computer equipment and medium |
| CN111150387A (en) * | 2020-01-15 | 2020-05-15 | 深圳市邦健科技有限公司 | Electrocardiogram template matching method and device, computer equipment and storage medium |
| CN111493856A (en) * | 2019-01-31 | 2020-08-07 | 深圳市理邦精密仪器股份有限公司 | Heart beat classification correction method, heart beat classification correction device, electrocardiogram checking system and storage medium |
| CN111616696A (en) * | 2020-05-20 | 2020-09-04 | 联想(北京)有限公司 | Electrocardiosignal detection method and device and storage medium |
| CN111772619A (en) * | 2020-06-19 | 2020-10-16 | 厦门纳龙科技有限公司 | Electrocardiogram heart beat identification method, terminal device and storage medium |
| CN111956203A (en) * | 2020-08-18 | 2020-11-20 | 广州视源电子科技股份有限公司 | Electrocardiosignal parameterization method, model training method, device, equipment and medium |
| CN113712567A (en) * | 2020-05-12 | 2021-11-30 | 深圳市科瑞康实业有限公司 | Method and device for generating interphase difference data sequence coefficients |
| CN113712525A (en) * | 2020-05-21 | 2021-11-30 | 深圳市理邦精密仪器股份有限公司 | Physiological parameter processing method and device and medical equipment |
| CN113768511A (en) * | 2020-06-04 | 2021-12-10 | 深圳市理邦精密仪器股份有限公司 | Physiological parameter detection method and electronic equipment |
| WO2022057180A1 (en) * | 2020-09-18 | 2022-03-24 | 深圳邦健生物医疗设备股份有限公司 | Heartbeat double-label analysis method, apparatus, device and medium |
| WO2022110524A1 (en) * | 2020-11-30 | 2022-06-02 | 深圳邦健生物医疗设备股份有限公司 | Method and device for electrocardiogram heartbeat data clustering, electronic device, and medium |
| WO2023158102A1 (en) * | 2022-02-18 | 2023-08-24 | 주식회사 뷰노 | Electrocardiogram delineation and result integration method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080002775A1 (en) * | 2006-06-30 | 2008-01-03 | Ricci Carlos A | Signal analysis employing matched wavelet |
| CN101856225A (en) * | 2010-06-30 | 2010-10-13 | 重庆大学 | Method for detecting R wave crest of electrocardiosignal |
| US20120068819A1 (en) * | 2006-04-05 | 2012-03-22 | Mcgrath William R | Systems and methods for remote long standoff biometric identification using microwave cardiac signals |
| CN102908135A (en) * | 2012-10-08 | 2013-02-06 | 中国科学院深圳先进技术研究院 | ECG diagnosis system and operating method of ECG diagnosis system |
| CN106778685A (en) * | 2017-01-12 | 2017-05-31 | 司马大大（北京）智能系统有限公司 | Electrocardiogram image-recognizing method, device and service terminal |
-
2017
- 2017-07-12 CN CN201710564424.4A patent/CN107358196B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120068819A1 (en) * | 2006-04-05 | 2012-03-22 | Mcgrath William R | Systems and methods for remote long standoff biometric identification using microwave cardiac signals |
| US20080002775A1 (en) * | 2006-06-30 | 2008-01-03 | Ricci Carlos A | Signal analysis employing matched wavelet |
| CN101856225A (en) * | 2010-06-30 | 2010-10-13 | 重庆大学 | Method for detecting R wave crest of electrocardiosignal |
| CN102908135A (en) * | 2012-10-08 | 2013-02-06 | 中国科学院深圳先进技术研究院 | ECG diagnosis system and operating method of ECG diagnosis system |
| CN106778685A (en) * | 2017-01-12 | 2017-05-31 | 司马大大（北京）智能系统有限公司 | Electrocardiogram image-recognizing method, device and service terminal |
Non-Patent Citations (2)
| Title |
|---|
| VESSELA KRASTEVA ET AL.: "Classification of supraventricular and ventricular beats by QRS template matching and decision tree", 《COMPUTING IN CARDIOLOGY 2014》 * |
| 季春林: "基于特征匹配的心电数据分类算法的研究与应用", 《中国优秀硕士学位论文全文数据库医药卫生科技辑》 * |
Cited By (48)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11234629B2 (en) | 2017-11-27 | 2022-02-01 | Shanghai Lepu CloudMed Co., Ltd. | Method and device for self-learning dynamic electrocardiography analysis employing artificial intelligence |
| CN107714023A (en) * | 2017-11-27 | 2018-02-23 | 乐普（北京）医疗器械股份有限公司 | Static ecg analysis method and apparatus based on artificial intelligence self study |
| CN107837082A (en) * | 2017-11-27 | 2018-03-27 | 乐普（北京）医疗器械股份有限公司 | Electrocardiogram automatic analysis method and device based on artificial intelligence self study |
| CN107951485A (en) * | 2017-11-27 | 2018-04-24 | 乐普（北京）医疗器械股份有限公司 | Ambulatory ECG analysis method and apparatus based on artificial intelligence self study |
| CN107981858A (en) * | 2017-11-27 | 2018-05-04 | 乐普（北京）医疗器械股份有限公司 | Electrocardiogram heartbeat automatic recognition classification method based on artificial intelligence |
| US11344243B2 (en) | 2017-11-27 | 2022-05-31 | Shanghai Lepu CloudMed Co., LTD | Artificial intelligence self-learning-based static electrocardiography analysis method and apparatus |
| US11564612B2 (en) | 2017-11-27 | 2023-01-31 | Shanghai Lepu CloudMed Co., LTD | Automatic recognition and classification method for electrocardiogram heartbeat based on artificial intelligence |
| WO2019100565A1 (en) * | 2017-11-27 | 2019-05-31 | 乐普（北京）医疗器械股份有限公司 | Method and device for self-learning dynamic electrocardiography analysis employing artificial intelligence |
| CN107951485B (en) * | 2017-11-27 | 2019-06-11 | 深圳市凯沃尔电子有限公司 | Ambulatory ECG analysis method and apparatus based on artificial intelligence self study |
| US11324455B2 (en) | 2017-11-27 | 2022-05-10 | Shanghai Lepu CloudMed Co., LTD | Artificial intelligence-based interference recognition method for electrocardiogram |
| CN107714023B (en) * | 2017-11-27 | 2020-09-01 | 上海优加利健康管理有限公司 | Static electrocardiogram analysis method and device based on artificial intelligence self-learning |
| CN107736888A (en) * | 2017-11-27 | 2018-02-27 | 中山大学 | A kind of method and system that real-time cardiac monitoring is carried out using wearable ecg equipment |
| WO2019100561A1 (en) * | 2017-11-27 | 2019-05-31 | 乐普（北京）医疗器械股份有限公司 | Artificial intelligence-based interference recognition method for electrocardiogram |
| WO2019100560A1 (en) * | 2017-11-27 | 2019-05-31 | 乐普（北京）医疗器械股份有限公司 | Artificial intelligence self-learning-based automatic electrocardiography analysis method and apparatus |
| US11529103B2 (en) | 2017-11-27 | 2022-12-20 | Shanghai Lepu CloudMed Co., LTD | Artificial intelligence self-learning-based automatic electrocardiography analysis method and apparatus |
| WO2019100566A1 (en) * | 2017-11-27 | 2019-05-31 | 乐普（北京）医疗器械股份有限公司 | Artificial intelligence self-learning-based static electrocardiography analysis method and apparatus |
| CN109948396B (en) * | 2017-12-20 | 2021-07-23 | 深圳市理邦精密仪器股份有限公司 | Heart beat classification method, heart beat classification device and electronic equipment |
| CN109948396A (en) * | 2017-12-20 | 2019-06-28 | 深圳市理邦精密仪器股份有限公司 | A kind of beat classification method, beat classification device and electronic equipment |
| CN108491769A (en) * | 2018-03-08 | 2018-09-04 | 四川大学 | Atrial fibrillation sorting technique based on phase between RR and multiple characteristic values |
| CN109077719A (en) * | 2018-07-05 | 2018-12-25 | 广州视源电子科技股份有限公司 | signal identification method, device, equipment and storage medium |
| CN108903935A (en) * | 2018-07-11 | 2018-11-30 | 上海夏先机电科技发展有限公司 | A kind of ventricular premature beat recognition methods, identifying system and electronic equipment |
| CN108937915A (en) * | 2018-07-24 | 2018-12-07 | 杭州质子科技有限公司 | The recognition methods of premature beat in a kind of dynamic ECG |
| CN108937915B (en) * | 2018-07-24 | 2021-02-19 | 杭州质子科技有限公司 | Method for identifying premature beat in dynamic electrocardiogram |
| CN109431488A (en) * | 2018-09-28 | 2019-03-08 | 上海优加利健康管理有限公司 | The recognition methods of typical data in a kind of electrocardiogram dynamic real-time analysis data |
| CN109303559A (en) * | 2018-11-01 | 2019-02-05 | 杭州质子科技有限公司 | A kind of dynamic ECG beat classification method promoting decision tree based on gradient |
| CN109303559B (en) * | 2018-11-01 | 2021-05-14 | 杭州质子科技有限公司 | Dynamic electrocardiogram and heartbeat classification method based on gradient boosting decision tree |
| CN109303561A (en) * | 2018-11-01 | 2019-02-05 | 杭州质子科技有限公司 | It is a kind of to clap the recognition methods clapped with the abnormal heart based on the artifact heart of misclassification and supervised learning |
| CN109871742B (en) * | 2018-12-29 | 2021-05-14 | 安徽心之声医疗科技有限公司 | Electrocardiosignal positioning method based on attention circulation neural network |
| CN109871742A (en) * | 2018-12-29 | 2019-06-11 | 安徽心之声医疗科技有限公司 | A kind of electrocardiosignal localization method based on attention Recognition with Recurrent Neural Network |
| CN111493856A (en) * | 2019-01-31 | 2020-08-07 | 深圳市理邦精密仪器股份有限公司 | Heart beat classification correction method, heart beat classification correction device, electrocardiogram checking system and storage medium |
| CN111493856B (en) * | 2019-01-31 | 2023-05-09 | 深圳市理邦精密仪器股份有限公司 | Correction method and device for heart beat classification, electrocardiogram checking system and storage medium |
| CN109691995A (en) * | 2019-01-31 | 2019-04-30 | 英菲泰克(天津)科技有限公司 | The arrhythmia cordis real-time data analysis system and method for single lead electrocardiogram |
| US12011277B2 (en) | 2019-12-27 | 2024-06-18 | Shenzhen Biocare Bio-Medical Equipment Co., Ltd. | RR interval ECG data distribution display method and device, computer equipment and medium |
| CN111053551A (en) * | 2019-12-27 | 2020-04-24 | 深圳邦健生物医疗设备股份有限公司 | RR interval electrocardio data distribution display method, device, computer equipment and medium |
| CN111150387A (en) * | 2020-01-15 | 2020-05-15 | 深圳市邦健科技有限公司 | Electrocardiogram template matching method and device, computer equipment and storage medium |
| CN113712567B (en) * | 2020-05-12 | 2023-09-01 | 深圳市科瑞康实业有限公司 | Method and device for generating heart beat interval difference value data sequence coefficient |
| CN113712567A (en) * | 2020-05-12 | 2021-11-30 | 深圳市科瑞康实业有限公司 | Method and device for generating interphase difference data sequence coefficients |
| CN111616696A (en) * | 2020-05-20 | 2020-09-04 | 联想(北京)有限公司 | Electrocardiosignal detection method and device and storage medium |
| CN113712525A (en) * | 2020-05-21 | 2021-11-30 | 深圳市理邦精密仪器股份有限公司 | Physiological parameter processing method and device and medical equipment |
| CN113712525B (en) * | 2020-05-21 | 2024-07-12 | 深圳市理邦精密仪器股份有限公司 | Physiological parameter processing method and device and medical equipment |
| CN113768511A (en) * | 2020-06-04 | 2021-12-10 | 深圳市理邦精密仪器股份有限公司 | Physiological parameter detection method and electronic equipment |
| CN113768511B (en) * | 2020-06-04 | 2023-09-22 | 深圳市理邦精密仪器股份有限公司 | Physiological parameter detection method and electronic equipment |
| CN111772619A (en) * | 2020-06-19 | 2020-10-16 | 厦门纳龙科技有限公司 | Electrocardiogram heart beat identification method, terminal device and storage medium |
| CN111956203A (en) * | 2020-08-18 | 2020-11-20 | 广州视源电子科技股份有限公司 | Electrocardiosignal parameterization method, model training method, device, equipment and medium |
| CN114271829A (en) * | 2020-09-18 | 2022-04-05 | 深圳邦健生物医疗设备股份有限公司 | Heart beat double-marker analysis method, device, equipment and medium |
| WO2022057180A1 (en) * | 2020-09-18 | 2022-03-24 | 深圳邦健生物医疗设备股份有限公司 | Heartbeat double-label analysis method, apparatus, device and medium |
| WO2022110524A1 (en) * | 2020-11-30 | 2022-06-02 | 深圳邦健生物医疗设备股份有限公司 | Method and device for electrocardiogram heartbeat data clustering, electronic device, and medium |
| WO2023158102A1 (en) * | 2022-02-18 | 2023-08-24 | 주식회사 뷰노 | Electrocardiogram delineation and result integration method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107358196B (en) | 2020-11-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107358196A (en) | A kind of sorting technique of heart beat type, device and electrocardiogram equipment | |
| CN107951485B (en) | Ambulatory ECG analysis method and apparatus based on artificial intelligence self study | |
| CN106214145B (en) | Electrocardiogram classification method based on deep learning algorithm | |
| US9042973B2 (en) | Apparatus and method for measuring physiological signal quality | |
| CN107837082A (en) | Electrocardiogram automatic analysis method and device based on artificial intelligence self study | |
| KR100954817B1 (en) | System and method for testing blood vessel's health and stress through signal analysis of pluse wave | |
| JP2005058766A (en) | Apparatus and method for detecting blood flow signal with removed dynamic noise, and stress testing apparatus using this | |
| CN102138789A (en) | Dynamic electrocardiogram and motion recording and analyzing system | |
| JP5996111B2 (en) | Evaluation device for myocardial damage based on current density fluctuation | |
| CN108814590A (en) | A kind of detection method and its ecg analysis method of Electrocardiograph QRS Wave group | |
| KR20100045521A (en) | Electrocardiogram derived apnoea/hypopnea index | |
| CN109288515B (en) | Periodicity monitoring method and device based on premature beat signal in wearable electrocardiosignal | |
| CN108814591A (en) | A kind of detection method and its ecg analysis method of Electrocardiograph QRS Wave group width | |
| CN106037720A (en) | Application method of hybrid continuous information analysis technology in medicine | |
| Ng et al. | Iterative method to detect atrial activations and measure cycle length from electrograms during atrial fibrillation | |
| Yang et al. | Heartbeat classification using discrete wavelet transform and kernel principal component analysis | |
| Ansari et al. | Noise detection in electrocardiography signal for robust heart rate variability analysis: A deep learning approach | |
| Sahoo | Analysis of ECG signal for Detection of Cardiac Arrhythmias | |
| Reddy et al. | Ecg signal characterization and correlation to heart abnormalities | |
| KR102150057B1 (en) | Method and apparatus for realtime detecting type of arrhythmia | |
| Firoozabadi et al. | Identification of exercise-induced ischemia using QRS slopes | |
| CN111839505A (en) | Atrial fibrillation classification method based on cardiovascular system electrical-mechanical activity information | |
| Sanamdikar et al. | Extraction of different features of ECG signal for detection of cardiac arrhythmias by using wavelet transformation Db 6 | |
| Gradl et al. | Temporal correction of detected R-peaks in ECG signals: A crucial step to improve QRS detection algorithms | |
| JP5933138B2 (en) | Apparatus and method for determining occurrence of QRS complex in ECG data |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20201110 Termination date: 20210712 |