CN103970975A - Electrocardio data processing method and electrocardio data processing system - Google Patents
Electrocardio data processing method and electrocardio data processing system Download PDFInfo
- Publication number
- CN103970975A CN103970975A CN201310041662.9A CN201310041662A CN103970975A CN 103970975 A CN103970975 A CN 103970975A CN 201310041662 A CN201310041662 A CN 201310041662A CN 103970975 A CN103970975 A CN 103970975A
- Authority
- CN
- China
- Prior art keywords
- ecg
- electrocardiogram
- data
- cluster
- 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
Abstract
The invention provides an electrocardio data processing method which comprises the following steps: collecting electrocardio data, processing the electrocardio data to obtain corresponding feature vectors, through the feature vectors, performing clustering analysis on the electrocardio data to obtain clustering population, and according to a template matched with the clustering population, obtaining a result corresponding to the electrocardio data. An electrocardio data processing system comprises an electrocardio recording device, a feature processing device, a cluster device and a cluster population processing device, wherein the electrocardio recording device is used for collecting the electrocardio data; the feature processing device is used for processing the electrocardio data to obtain the corresponding corresponding feature vectors; the cluster device is used for performing clustering analysis on the electrocardio data to obtain the clustering population through the feature vectors; the cluster population processing device is used for obtaining the result corresponding to the electrocardio data according to the template matched with the clustering population. With adoption of the electrocardio data processing method and the electrocardio data processing system, the accuracy of analysis on the electrocardio data can be improved.
Description
Technical field
The present invention relates to computer technology, particularly relate to a kind of electrocardiogram (ECG) data disposal route and system.
Background technology
Heart disease is one of main diseases kind threatening human life's health, conventionally only has and early finds that early treatment just has larger may healing.Electrocardiogram (ECG) data has reflected the function of heart, can utilize electrocardiogram (ECG) data to realize the diagnosis of heart disease.The collection of electrocardiogram (ECG) data mainly realizes by ordinary electrocardiogram and two kinds of modes of Holter, wherein, ordinary electrocardiogram (Electrocardiogram, be called for short ECG) diagnosis be in modern medicine for one of important tool of Diagnosing Cardiac disease, be widely used in clinical medicine.
Holter is a kind of method that can long-time continuous record and compile the cardiogram variation under movable and rest state of analyst's systemic heart, compare with ordinary electrocardiogram, Holter can be in 24 hours the nearly electrocardiosignal of 100,000 left and right of continuous recording, therefore, can effectively improve the non-standing rhythm of the heart is played the part of, especially can greatly improve the recall rate of arrhythmia and of short duration myocardial ischemic attacks.
But the electrocardiogram (ECG) data that Holter records is that repeated and redundant, data volume are large mostly, only relying on electrocardio doctor or paramedic to observe for a long time cardiogram can diagnose, and tends to cause undetected because of the dispersion of asthenopia or notice.
Along with the development of computer technology, adopt more and more ECG automatic analysis system to analyze electrocardiogram (ECG) data, to avoid electrocardio doctor or paramedic to observe for a long time cardiogram.
But, this traditional ECG automatic analysis system is difficult to realize large-scale clinical practice, ecg wave form individual difference is larger, global classification device single in ECG automatic analysis system does not have universality, that is to say, single global classification device cannot all possess consistent general high-accuracy to first patient's electrocardiogram (ECG) data, and then causes the inaccurate analysis of electrocardiogram (ECG) data.
Summary of the invention
Based on this, be necessary to provide a kind of electrocardiogram (ECG) data disposal route of the accuracy that can improve electrocardiogram (ECG) data analysis.
In addition, be also necessary to provide a kind of electrocardiogram (ECG) data disposal system of the accuracy that can improve electrocardiogram (ECG) data analysis.
A kind of electrocardiogram (ECG) data disposal route, comprises the steps:
Gather electrocardiogram (ECG) data;
Process described electrocardiogram (ECG) data and obtain corresponding proper vector;
By described proper vector, described electrocardiogram (ECG) data is carried out to cluster analysis and obtain cluster population;
Obtain according to the template of described cluster population coupling the result that described electrocardiogram (ECG) data is corresponding.
In an embodiment, the step that the described electrocardiogram (ECG) data of described processing obtains corresponding proper vector comprises therein:
Electrocardiogram (ECG) data described in pre-service;
Respectively described pretreated electrocardiogram (ECG) data is carried out to the solving of wavelet transformation characteristic of field, the calculating of RR interval feature and the calculating of T ripple fuzzy characteristics, to obtain corresponding eigenwert;
By described eigenwert composition proper vector.
In an embodiment, the step that the described template according to described cluster population coupling obtains the result that described electrocardiogram (ECG) data is corresponding comprises therein:
Each cluster population is mated one by one with template, obtain the matching ratio between described cluster population and described template;
Judge whether described matching ratio is greater than threshold value, if so,
Obtain the result corresponding with described template.
In an embodiment, the step that the described template according to described cluster population coupling obtains the result that described electrocardiogram (ECG) data is corresponding also comprises therein:
If the matching ratio determining between each cluster population and all templates is not greater than threshold value, described cluster population is screened and obtains representative electrocardio beat, and send described representative electrocardio beat.
Therein in an embodiment, described described cluster population screened and obtains representative electrocardio beat, and also comprise after sending the step of described representative electrocardio beat:
Receive according to described representative electrocardio beat feedack;
Process described cluster population according to described feedack.
Therein in an embodiment, the described step of processing described cluster population according to described feedack comprises:
Judge the cluster instruction of whether attaching most importance to of described feedack, if so, adjust clustering parameter, the corresponding electrocardiogram (ECG) data of described cluster population is carried out to cluster again, the cluster population of the class that obtains meeting again, if not,
Extract the template recording in described feedack, and show the corresponding result of template of described extraction.
A kind of electrocardiogram (ECG) data disposal system, comprising:
Cardiac electric recording apparatus, for gathering electrocardiogram (ECG) data;
Characteristic processing device, obtains corresponding proper vector for the treatment of described electrocardiogram (ECG) data;
Clustering apparatus, obtains cluster population for described electrocardiogram (ECG) data being carried out to cluster analysis by described proper vector;
Cluster kind cluster treating device, for obtaining according to the template of described cluster population coupling the result that described electrocardiogram (ECG) data is corresponding.
In an embodiment, described characteristic processing device comprises therein:
Pretreatment module, for electrocardiogram (ECG) data described in pre-service;
Eigenwert computing module, for respectively described pretreated electrocardiogram (ECG) data being carried out to the solving of wavelet transformation characteristic of field, the calculating of RR interval feature and the calculating of T ripple fuzzy characteristics, to obtain corresponding eigenwert;
Vector composition module, for forming proper vector by described eigenwert.
In an embodiment, described cluster kind cluster treating device comprises therein:
Matching module, for each cluster population is mated one by one with template, obtains the matching ratio between described cluster population and described template;
Matching judgment module, for judging whether described matching ratio is greater than threshold value, if so, advise fate acquisition module;
Described result acquisition module is for obtaining the result corresponding with described template.
In an embodiment, described cluster kind cluster treating device also comprises screening module therein;
If the matching ratio determining between each cluster population and all templates is not greater than threshold value, described screening module is screened and is obtained representative electrocardio beat described cluster population, and sends described representative electrocardio beat.
In an embodiment, described system also comprises therein:
Information receiver, for receiving according to described representative electrocardio beat feedack;
Feedback processing device, for processing described cluster population according to described feedack.
In an embodiment, described feedback processing device also comprises therein:
Signal judgement module, for judging the cluster instruction of whether attaching most importance to of described feedack, if so, notifies reunion generic module, if not, notifies extraction module;
Described reunion generic module is used for adjusting clustering parameter, and the corresponding electrocardiogram (ECG) data of described cluster population is carried out to cluster again, the cluster population of the class that obtains meeting again;
Described extraction module is used for extracting the template that described feedack records, and shows the corresponding result of template of described extraction.
Above-mentioned electrocardiogram (ECG) data disposal route and system, the electrocardiogram (ECG) data collecting is processed, to obtain corresponding proper vector, by proper vector, electrocardiogram (ECG) data is carried out to cluster analysis to obtain cluster population, realized the classification of electrocardiogram (ECG) data, and then obtain the corresponding result of this electrocardiogram (ECG) data according to the template of cluster population coupling, compared with prior art, a large amount of electrocardiogram (ECG) datas has progressively been realized to refinement the analysis of electrocardiogram (ECG) data by classification and template matches, take into full account the individual difference between electrocardiogram (ECG) data, and then obtain result accurately, accuracy and universality that electrocardiogram (ECG) data is analyzed are improved widely, be conducive to realize long cardiac monitoring.
Brief description of the drawings
Fig. 1 is the process flow diagram of the electrocardiogram (ECG) data disposal route of an embodiment;
Fig. 2 is the method flow diagram of processing electrocardiogram (ECG) data in Fig. 1 and obtain corresponding proper vector;
Fig. 3 is the method flow diagram that obtains the result that electrocardiogram (ECG) data is corresponding in Fig. 1 according to the template of cluster population coupling;
Fig. 4 is the process flow diagram of the electrocardiogram (ECG) data disposal route of another embodiment;
Fig. 5 is the method flow diagram of processing cluster population in Fig. 4 according to feedack;
Fig. 6 is the structural representation of the electrocardiogram (ECG) data disposal system of an embodiment;
Fig. 7 is the structural representation of characteristic processing device in Fig. 6;
Fig. 8 is the structural representation of cluster kind cluster treating device in Fig. 6;
Fig. 9 is the structural representation of the electrocardiogram (ECG) data disposal system of another embodiment;
Figure 10 is the structural representation of feedback processing device in Fig. 9.
Embodiment
As shown in Figure 1, in one embodiment, a kind of electrocardiogram (ECG) data disposal route, comprises the steps:
Step S110, gathers electrocardiogram (ECG) data.
In the present embodiment, after user has completed the connection of cardiac diagnosis lead, gather user's electrocardiogram (ECG) data, and the electrocardiogram (ECG) data gathering is stored in storer, wherein, this storer is preferably the SD card that large capacity, storage and transmission speed are fast, volume is little (Secure Digital Memory Card, safe digital card).
Further, for realizing the cardiac electric recording apparatus of electrocardio-data collection, for example, Dynamic Cardiogram Record, will adopt multi-lead structure, as, two lead, five lead or 12 lead.In preferred embodiment, also employing bandpass filter and notch filter are carried out to filtering to electrocardiogram (ECG) data, and then eliminate the interference noise in electrocardiogram (ECG) data.
Step S130, processes electrocardiogram (ECG) data and obtains corresponding proper vector.
In the present embodiment, proper vector will accurately and briefly show the feature of electrocardiogram (ECG) data, therefore, need to process to obtain corresponding proper vector to electrocardiogram (ECG) data, to characterize exactly electrocardiogram (ECG) data by proper vector.
Step S150, carries out cluster analysis by proper vector to electrocardiogram (ECG) data and obtains cluster population.
In the present embodiment, application characteristic vector is realized the accurate classification of electrocardiogram (ECG) data, wherein, the clustering method using can be K means clustering method, K central point clustering method or expect maximum (EM) clustering method etc., and then obtain cluster population and the corresponding population information of this cluster population.Cluster population will comprise in other electrocardiogram (ECG) data of same class.
Step S170, obtains according to the template of cluster population coupling the result that electrocardiogram (ECG) data is corresponding.
In the present embodiment, pre-stored multiple templates, this template is the electrocardiogram (ECG) data sample extracting in advance, result that each template is all corresponding, can know the corresponding situation of current electrocardiogram (ECG) data by this result, for example, outcome record assessment result and the solution to electrocardiogram (ECG) data, and then make user can know the situation of current heart and alleviate the solution of current heart by result.From pre-stored multiple templates, obtain the template of mating with cluster groupy phase,
As shown in Figure 2, in one embodiment, the detailed process of above-mentioned steps S130 is:
Step S131, pre-service electrocardiogram (ECG) data.
In the present embodiment, read the electrocardiogram (ECG) data of collection, electrocardiogram (ECG) data is carried out to filtering.Concrete, can realize by bandpass filter, for example, this bandpass filter can be that Hi-pass filter and the cutoff frequency that 0.1Hz, stopband attenuation are 30dB is that the low-pass filter that 35Hz, stopband attenuation are 80dB forms by cutoff frequency.
Further, after having completed the filtering of electrocardiogram (ECG) data, also will carry out wavelet decomposition to filtered electrocardiogram (ECG) data.In a preferred embodiment, the wavelet decomposition of carrying out is four layers of wavelet decomposition, and wherein, female small echo adopts the western small echo of the many shellfishes of db4(, N=4), and adopt porous algorithm (Algorithm a ' trous).
Step S133, carries out the solving of wavelet transformation characteristic of field, the calculating of RR interval feature and the calculating of T ripple fuzzy characteristics to pretreated electrocardiogram (ECG) data respectively, to obtain corresponding eigenwert.
In the present embodiment, it is mainly the statistical information that solves the 4th yardstick of electrocardiogram (ECG) data wavelet decomposition that pretreated electrocardiogram (ECG) data is carried out to solving of wavelet transformation characteristic of field, to calculate following characteristics value:
C
2x(k)＝E{x(n)x(n+k)}
C
3x(k,l)＝E{x(n)x(n+k)x(n+l)}
C
4x(k,l,m)＝E{x(n)x(n+k)x(n+l)x(n+m)}
-C
2x(k)C
2x(m-l)-C
2x(l)C
2x(m-k)-C
2x(m)C
2x(l-k)
Wherein, x (n) represents the component of the 4th layer of electrocardiogram (ECG) data wavelet decomposition, and E represents to ask expectation, k, and l, m, represents time delay.
The calculating of RR interval feature is to carry out according to the R ripple position of electrocardiogram (ECG) data, before calculating, should carry out the detection of R ripple to pretreated electrocardiogram (ECG) data.Concrete, because the R wave energy of electrocardiogram (ECG) data mainly concentrates on the responsive yardstick of wavelet decomposition, therefore, to extract R wave-wave peak point at the 3rd yardstick, according to the Wavelet Detection principle of singular point, R ripple is corresponding to the extreme value pair on the 3rd yardstick, and R wave-wave peak dot is by corresponding to the right zero crossing of extreme value.
Detect the position that has obtained R ripple by R ripple, i.e. extreme value pair on the 3rd yardstick, now, will define following formula:
RR[i]＝R[i]-R[i-1]
diffRR[i]＝RR[i]-RR[i-1]
Calculate RR[i according to above-mentioned formula], RR[i-1], diffRRmean[i, 1], RRmean[i, 5] and RRmean[i, 20] etc. eigenwert, and then set it as the Partial Feature of cluster analysis.
Wherein R[i] represent in electrocardiogram (ECG) data, i the time that R ripple occurs, RR[i] represent the Time Intervals of i R ripple and i-1 R ripple, diffRR[i] represent the variation of i RR interval, diffRRmean[i, k] represent the mean value of k RR interval changing value after i R wavefront, RRmean[i, k] mean value of k RR interval after i R wavefront of expression.
Carrying out, between T mode paste feature calculation, first carrying out the detection of T ripple.In electrocardiogram (ECG) data, the energy of T ripple mainly concentrates on the 4th yardstick of wavelet decomposition, therefore, detect the R ripple position obtaining according to R ripple, on wavelet decomposition the 4th yardstick according to R ripple position in corresponding search window search respectively corresponding T ripple position and T wave-wave peak extreme value to and zero crossing, for example, this search window can be from current QRS ripple terminal, finishes to the distance of the RR interval of 0.5 times.
Calculate the distance spacing interval_RT of R ripple position and T ripple position, and detect T waveform state, corresponding T ripple inversion, normal T ripple and existing inversion also have the inversion of the T ripple of forward, the normal and two-way fuzzy quantity that is definition respectively.
Step S135, forms proper vector by eigenwert.
In the present embodiment, by solve wavelet transformation characteristic of field gained to eigenwert, calculate eigenwert that RR interval feature obtains and the proper vector of calculating eigenvalue cluster that T ripple fuzzy characteristics obtains and become electrocardiogram (ECG) data.
As shown in Figure 3, in one embodiment, the detailed process of above-mentioned steps S170 is:
Step S171, mates each cluster population one by one with template, obtains the matching ratio between cluster population and template.
In the present embodiment, the cluster population that cluster analysis is obtained is carried out to template matches one by one, the template of being mated to obtain each cluster population.
Concrete, each cluster population is mated one by one with multiple templates of pre-stored, to obtain the matching ratio between this cluster population and each template.
Step S173, judges whether matching ratio is greater than threshold value, if so, enters step S175, if not,, enters step S177.
In the present embodiment, if the matching ratio determining between cluster population and a certain template is greater than threshold value, illustrate that this template is the template of mating with cluster groupy phase, if the matching ratio determining between cluster population and all templates is not all greater than threshold value, illustrate that current None-identified goes out the template that this cluster population mates.
Step S175, obtains the result corresponding with this template.
In the present embodiment, because each template all has corresponding result, therefore, determining after matching ratio between cluster population and a certain template is greater than threshold value, to directly obtain the corresponding result of this template, this result is the current residing situation of electrocardiogram (ECG) data that cluster population comprises.
Step S177, screens and obtains representative electrocardio beat cluster population, and sends representative electrocardio beat.
In the present embodiment, sort to obtain the center of cluster population according to eigenwert, nearest several electrocardio beats of the central point of this cluster population of selected distance and apart from the central point of this cluster population several electrocardio beats farthest in the electrocardiogram (ECG) data comprising of cluster population, these electrocardio beats of choosing are representative electrocardio beat.
As shown in Figure 4, in one embodiment, after above-mentioned steps S177, also comprise:
Step S210, receives according to representative electrocardio beat feedack.
In the present embodiment, after the representative electrocardio beat that screening is obtained sends with the form of packet, remote subscriber is after carrying out identification by input identify label and password, to receive this packet, and resolve this packet with obtain cannot with the cluster population of template matches, now, by the information of obtaining cluster population that remote subscriber obtains according to parsing and inputting, and this information of packing, and feed back this information by internet in the mode of online or off-line, this information comprises it can being the instruction of reunion class, also can be the information of specifying a certain template for cluster population.
Step S230, processes cluster population according to feedack.
In the present embodiment, according to feedack to cannot with meet again class or obtain the result corresponding with the template of specifying of the cluster population of template matches, to make electrocardiogram (ECG) data also can obtain result accurately in the situation that template matches cannot be processed, and then constantly study positive template matching process.
As shown in Figure 5, in one embodiment, the detailed process of above-mentioned steps S230 is:
Step S231, judges the feedack cluster instruction of whether attaching most importance to, and if so, enters step S233, if not, enters step S235.
In the present embodiment, if determine the feedack cluster instruction of attaching most importance to, cannot carry out cluster again with the cluster population of template matches to this, if when determining feedack and not being the instruction of reunion class, illustrate and in feedack, recorded the corresponding template of cluster population.
Step S233, adjusts clustering parameter, and the corresponding electrocardiogram (ECG) data of cluster population is carried out to cluster again, the cluster population of the class that obtains meeting again.
In the present embodiment, the clustering parameter of adjusting can be the population number of increase cluster etc., to make the reunion class process on backstage more accurate, after having completed again cluster, the cluster population of counterweight cluster described above is carried out to template matches, with the template that obtains mating with cluster population, and then obtain corresponding result, do not repeat them here.
Step S235, extracts the template recording in feedack, and shows the corresponding result of template of extracting.
In the present embodiment, the result of cluster population is shown, to realize the information interaction between user.Concrete, display mode can be the mode of screen.
As shown in Figure 6, in one embodiment, a kind of electrocardiogram (ECG) data disposal system comprises cardiac electric recording apparatus 110, characteristic processing device 130, clustering apparatus 150 and cluster kind cluster treating device 170.
Cardiac electric recording apparatus 110, for gathering electrocardiogram (ECG) data.
In the present embodiment, after user has completed the connection of cardiac diagnosis lead, cardiac electric recording apparatus 110 gathers user's electrocardiogram (ECG) data, and the electrocardiogram (ECG) data gathering is stored in storer, wherein, this storer is preferably the SD card that large capacity, storage and transmission speed are fast, volume is little (Secure Digital Memory Card, safe digital card).
Further, for realizing the cardiac electric recording apparatus 110 of electrocardio-data collection, for example, Dynamic Cardiogram Record, will adopt multi-lead structure, as, two lead, five lead or 12 lead.In preferred embodiment, cardiac electric recording apparatus 110 has comprised bandpass filter and notch filter, so that electrocardiogram (ECG) data is carried out to filtering, and then eliminates the interference noise in electrocardiogram (ECG) data.
Characteristic processing device 130, obtains corresponding proper vector for the treatment of electrocardiogram (ECG) data.
In the present embodiment, proper vector will accurately and briefly show the feature of electrocardiogram (ECG) data, and therefore, characteristics of needs treating apparatus 130 processes to obtain corresponding proper vector to electrocardiogram (ECG) data, to characterize exactly electrocardiogram (ECG) data by proper vector.
Clustering apparatus 150, obtains cluster population for electrocardiogram (ECG) data being carried out to cluster analysis by proper vector.
In the present embodiment, clustering apparatus 150 application characteristic vectors are realized the accurate classification of electrocardiogram (ECG) data, wherein, the clustering method using can be K means clustering method, K central point clustering method or expect maximum (EM) clustering method etc., and then obtain cluster population and the corresponding population information of this cluster population.Cluster population will comprise in other electrocardiogram (ECG) data of same class.
Cluster kind cluster treating device 170, for obtaining according to the template of cluster population coupling the result that electrocardiogram (ECG) data is corresponding.
In the present embodiment, pre-stored multiple templates, this template is the electrocardiogram (ECG) data sample extracting in advance, result that each template is all corresponding, can know the corresponding situation of current electrocardiogram (ECG) data by this result, for example, outcome record assessment result and the solution to electrocardiogram (ECG) data, and then make user can know the situation of current heart and alleviate the solution of current heart by result.From pre-stored multiple templates, obtain the template of mating with cluster groupy phase,
As shown in Figure 7, in one embodiment, above-mentioned characteristic processing device 130 comprises pretreatment module 131, eigenwert computing module 133 and vector composition module 135.
Pretreatment module 131, for pre-service electrocardiogram (ECG) data.
In the present embodiment, pretreatment module 131 reads the electrocardiogram (ECG) data of collection, and electrocardiogram (ECG) data is carried out to filtering.Concrete, can realize by bandpass filter, for example, this bandpass filter can be that Hi-pass filter and the cutoff frequency that 0.1Hz, stopband attenuation are 30dB is that the low-pass filter that 35Hz, stopband attenuation are 80dB forms by cutoff frequency.
Further, after having completed the filtering of electrocardiogram (ECG) data, pretreatment module 131 also will be carried out wavelet decomposition to filtered electrocardiogram (ECG) data.In a preferred embodiment, the wavelet decomposition of carrying out is four layers of wavelet decomposition, and wherein, female small echo adopts the western small echo of the many shellfishes of db4(, N=4), and employing porous algorithm (Algorithma ' trous).
Eigenwert computing module 133, for respectively pretreated electrocardiogram (ECG) data being carried out to the solving of wavelet transformation characteristic of field, calculating that RR asks phase feature and the calculating of T ripple fuzzy characteristics, to obtain corresponding eigenwert.
In the present embodiment, it is mainly the statistical information that solves the 4th yardstick of electrocardiogram (ECG) data wavelet decomposition that eigenwert computing module 133 carries out solving of wavelet transformation characteristic of field to pretreated electrocardiogram (ECG) data, to calculate following characteristics value:
C
2x(k)＝E{x(n)x(n+k)}
C
3x(k,l)＝E{x(n)x(n+k)x(n+l)}
C
4x(k,l,m)＝E{x(n)x(n+k)x(n+l)x(n+m)}
-C
2x(k)C
2x(m-l)-C
2x(l)C
2x(m-k)-C
2x(m)C
2x(l-k)
Wherein, x (n) represents the component of the 4th layer of electrocardiogram (ECG) data wavelet decomposition, and E represents to ask expectation, k, and l, m, represents time delay.
In eigenwert computing module 133, the calculating of RR interval feature is to carry out according to the R ripple position of electrocardiogram (ECG) data, and before calculating, eigenwert computing module 133 should carry out the detection of R ripple to pretreated electrocardiogram (ECG) data.Concrete, because the R wave energy of electrocardiogram (ECG) data mainly concentrates on the responsive yardstick of wavelet decomposition, therefore, eigenwert computing module 133 will extract R wave-wave peak point at the 3rd yardstick, according to the Wavelet Detection principle of singular point, R ripple is corresponding to the extreme value pair on the 3rd yardstick, and R wave-wave peak dot is by corresponding to the right zero crossing of extreme value.
Eigenwert computing module 133 detects the position that has obtained R ripple by R ripple, i.e. extreme value pair on the 3rd yardstick, now, will define following formula:
RR[i]＝R[i]-R[i-1]
diffRR[i]＝RR[i]-RR[i-1]
Calculate RR[i according to above-mentioned formula], RR[i-1], diffRRmean[i, 1], RRmean[i, 5] and RRmean[i, 20] etc. eigenwert, and then set it as the Partial Feature of cluster analysis.
Wherein R[i] represent in electrocardiogram (ECG) data, i the time that R ripple occurs, RR[i] represent the Time Intervals of i R ripple and i-1 R ripple, diffRR[i] represent the variation of i RR interval, diffRRmean[i, k] represent the mean value of k RR interval changing value after i R wavefront, RRmean[i, k] mean value of k RR interval after i R wavefront of expression.
Carrying out between T mode paste feature calculation, first vector composition module 135 will carry out the detection of T ripple.In electrocardiogram (ECG) data, the energy of T ripple mainly concentrates on the 4th yardstick of wavelet decomposition, therefore, vector composition module 135 detects the R ripple position obtaining according to R ripple, on wavelet decomposition the 4th yardstick according to R ripple position in corresponding search window search respectively corresponding T ripple position and T wave-wave peak extreme value to and zero crossing, for example, this search window can be from current QRS ripple terminal, finishes to the distance of the RR interval of 0.5 times.
Vector composition module 135 is calculated the distance spacing interval_RT of R ripple position and T ripple position, and detects T waveform state, and corresponding T ripple inversion, normal T ripple and existing inversion also have the inversion of the T ripple of forward, the normal and two-way fuzzy quantity that is definition respectively.
Vector composition module 135, for forming proper vector by eigenwert.
In the present embodiment, vector composition module 135 by solve wavelet transformation characteristic of field gained to eigenwert, calculate eigenwert that RR interval feature obtains and the proper vector of calculating eigenvalue cluster that T ripple fuzzy characteristics obtains and become electrocardiogram (ECG) data.
As shown in Figure 8, in one embodiment, above-mentioned cluster kind cluster treating device 170 comprises matching module 171, matching judgment module 173, result acquisition module 175 and screening module 177.
Matching module 171, for each cluster population is mated one by one with template, obtains the matching ratio between cluster population and template.
In the present embodiment, the cluster population that cluster analysis is obtained is carried out one by one template matches by matching module 171, the template of being mated to obtain each cluster population.
Concrete, matching module 171 mates each cluster population one by one with multiple templates of pre-stored, to obtain the matching ratio between this cluster population and each template.
Matching judgment module 173, for judging whether matching ratio is greater than threshold value, if so, advise fate acquisition module 175, if not, notice screening module 177.
In the present embodiment, if the matching ratio that matching judgment module 173 determines between cluster population and a certain template is greater than threshold value, illustrate that this template is the template of mating with cluster groupy phase, if the matching ratio that matching judgment module 173 determines between cluster population and all templates is not all greater than threshold value, illustrate that current None-identified goes out the template that this cluster population mates.
Result acquisition module 175, for obtaining the result corresponding with this template.
In the present embodiment, because each template all has corresponding result, therefore, determining after matching ratio between cluster population and a certain template is greater than threshold value, result acquisition module 175 will directly obtain the corresponding result of this template, and this result is the current residing situation of electrocardiogram (ECG) data that cluster population comprises.
Screening module 177, for cluster population is screened and obtains representative electrocardio beat, and sends representative electrocardio beat.
In the present embodiment, screening module 177 sorts to obtain the center of cluster population according to eigenwert, nearest several electrocardio beats of the central point of this cluster population of selected distance and apart from the central point of this cluster population several electrocardio beats farthest in the electrocardiogram (ECG) data comprising of cluster population, these electrocardio beats of choosing are representative electrocardio beat.
As shown in Figure 9, in one embodiment, above-mentioned electrocardiogram (ECG) data disposal system also comprises information receiver 210 and feedback processing device 230.
Information receiver 210, for receiving according to representative electrocardio beat feedack.
In the present embodiment, above-mentioned electrocardiogram (ECG) data disposal system also comprises servicing unit, the representative electrocardio beat that screening is obtained sends out the servicing unit to far-end with the form of packet, now remote subscriber by servicing unit after carrying out identification by input identify label and password, to receive this packet, and resolve this packet with obtain cannot with the cluster population of template matches, now, by the information of obtaining cluster population that remote subscriber obtains according to parsing and inputting, and this information of packing, and feed back this information by internet in the mode of online or off-line, this information comprises it can being the instruction of reunion class, also can be the information of specifying a certain template for cluster population.
Feedback processing device 230, for processing cluster population according to feedack.
In the present embodiment, feedback processing device 230 according to feedack to cannot with meet again class or obtain the result corresponding with the template of specifying of the cluster population of template matches, to make electrocardiogram (ECG) data also can obtain result accurately in the situation that template matches cannot be processed, and then constantly study positive template matching process.
As shown in figure 10, in one embodiment, above-mentioned feedback processing device 230 comprises signal judgement module 231, reunion generic module 233 and extraction module 235.
Signal judgement module 231, for judging the feedack cluster instruction of whether attaching most importance to, if so, notifies reunion generic module 233, if not, notifies extraction module 235.
In the present embodiment, if signal judgement module 231 determines the feedack cluster instruction of attaching most importance to, cannot carry out cluster again with the cluster population of template matches to this, if when determining feedack and not being the instruction of reunion class, illustrate and in feedack, recorded the corresponding template of cluster population.
Reunion generic module 233, for adjusting clustering parameter, carries out cluster again to the corresponding electrocardiogram (ECG) data of cluster population, the cluster population of the class that obtains meeting again.
In the present embodiment, the clustering parameter that reunion generic module 233 is adjusted can be the population number of increase cluster etc., to make the reunion class process on backstage more accurate, after having completed again cluster, the cluster population of counterweight cluster described above is carried out to template matches, with the template that obtains mating with cluster population, and then obtain corresponding result, do not repeat them here.
Extraction module 235, the template recording for extracting feedack, and show the corresponding result of template of extracting.
In the present embodiment, the result of cluster population is shown, to realize the information interaction between user.Concrete, display mode can be the mode of screen.
In above-mentioned electrocardiogram (ECG) data disposal system, characteristic processing device, clustering apparatus, cluster kind cluster treating device, information receiver and feedback processing installation composition user side ecg analysis system, run in user's computer, and realize information interaction with the servicing unit of far-end, multiple user side ecg analysis systems and servicing unit carry out alternately.
Above-mentioned electrocardiogram (ECG) data disposal route and system, the electrocardiogram (ECG) data collecting is processed, to obtain corresponding proper vector, by proper vector, electrocardiogram (ECG) data is carried out to cluster analysis to obtain cluster population, realized the classification of electrocardiogram (ECG) data, and then obtain the corresponding result of this electrocardiogram (ECG) data according to the template of cluster population coupling, compared with prior art, a large amount of electrocardiogram (ECG) datas has progressively been realized to refinement the analysis of electrocardiogram (ECG) data by classification and template matches, and then obtain result accurately, accuracy and universality that electrocardiogram (ECG) data is analyzed are improved widely, be conducive to realize long cardiac monitoring.
One of ordinary skill in the art will appreciate that all or part of flow process realizing in above-described embodiment method, can carry out the hardware that instruction is relevant by computer program to complete, described program can be stored in a computer read/write memory medium, this program, in the time carrying out, can comprise as the flow process of the embodiment of above-mentioned each side method.Wherein, described storage medium can be magnetic disc, CD, read-only store-memory body (Read-Only Memory, ROM) or random store-memory body (Random Access Memory, RAM) etc.
The above embodiment has only expressed several embodiment of the present invention, and it describes comparatively concrete and detailed, but can not therefore be interpreted as the restriction to the scope of the claims of the present invention.It should be pointed out that for the person of ordinary skill of the art, without departing from the inventive concept of the premise, can also make some distortion and improvement, these all belong to protection scope of the present invention.Therefore, the protection domain of patent of the present invention should be as the criterion with claims.
Claims (12)
1. an electrocardiogram (ECG) data disposal route, comprises the steps:
Gather electrocardiogram (ECG) data;
Process described electrocardiogram (ECG) data and obtain corresponding proper vector;
By described proper vector, described electrocardiogram (ECG) data is carried out to cluster analysis and obtain cluster population;
Obtain according to the template of described cluster population coupling the result that described electrocardiogram (ECG) data is corresponding.
2. electrocardiogram (ECG) data disposal route according to claim 1, is characterized in that, the step that the described electrocardiogram (ECG) data of described processing obtains corresponding proper vector comprises:
Electrocardiogram (ECG) data described in pre-service;
Respectively described pretreated electrocardiogram (ECG) data is carried out to the solving of wavelet transformation characteristic of field, the calculating of RR interval feature and the calculating of T ripple fuzzy characteristics, to obtain corresponding eigenwert;
By described eigenwert composition proper vector.
3. electrocardiogram (ECG) data disposal route according to claim 1, is characterized in that, the step that the described template according to described cluster population coupling obtains the result that described electrocardiogram (ECG) data is corresponding comprises:
Each cluster population is mated one by one with template, obtain the matching ratio between described cluster population and described template;
Judge whether described matching ratio is greater than threshold value, if so,
Obtain the result corresponding with described template.
4. electrocardiogram (ECG) data disposal route according to claim 3, is characterized in that, the step that the described template according to described cluster population coupling obtains the result that described electrocardiogram (ECG) data is corresponding also comprises:
If the matching ratio determining between each cluster population and all templates is not greater than threshold value, described cluster population is screened and obtains representative electrocardio beat, and send described representative electrocardio beat.
5. electrocardiogram (ECG) data disposal route according to claim 4, is characterized in that, described described cluster population is screened and obtains representative electrocardio beat, and also comprise after sending the step of described representative electrocardio beat:
Receive according to described representative electrocardio beat feedack;
Process described cluster population according to described feedack.
6. electrocardiogram (ECG) data disposal route according to claim 5, is characterized in that, the described step of processing described cluster population according to described feedack comprises:
Judge the cluster instruction of whether attaching most importance to of described feedack, if so, adjust clustering parameter, the corresponding electrocardiogram (ECG) data of described cluster population is carried out to cluster again, the cluster population of the class that obtains meeting again, if not,
Extract the template recording in described feedack, and show the corresponding result of template of described extraction.
7. an electrocardiogram (ECG) data disposal system, is characterized in that, comprising:
Cardiac electric recording apparatus, for gathering electrocardiogram (ECG) data;
Characteristic processing device, obtains corresponding proper vector for the treatment of described electrocardiogram (ECG) data;
Clustering apparatus, obtains cluster population for described electrocardiogram (ECG) data being carried out to cluster analysis by described proper vector;
Cluster kind cluster treating device, for obtaining according to the template of described cluster population coupling the result that described electrocardiogram (ECG) data is corresponding.
8. electrocardiogram (ECG) data disposal system according to claim 7, is characterized in that, described characteristic processing device comprises:
Pretreatment module, for electrocardiogram (ECG) data described in pre-service;
Eigenwert computing module, for respectively described pretreated electrocardiogram (ECG) data being carried out to the solving of wavelet transformation characteristic of field, the calculating of RR interval feature and the calculating of T ripple fuzzy characteristics, to obtain corresponding eigenwert;
Vector composition module, for forming proper vector by described eigenwert.
9. electrocardiogram (ECG) data disposal system according to claim 7, is characterized in that, described cluster kind cluster treating device comprises:
Matching module, for each cluster population is mated one by one with template, obtains the matching ratio between described cluster population and described template;
Matching judgment module, for judging whether described matching ratio is greater than threshold value, if so, advise fate acquisition module;
Described result acquisition module is for obtaining the result corresponding with described template.
10. electrocardiogram (ECG) data disposal system according to claim 9, is characterized in that, described cluster kind cluster treating device also comprises screening module;
If the matching ratio determining between each cluster population and all templates is not greater than threshold value, described screening module is screened and is obtained representative electrocardio beat described cluster population, and sends described representative electrocardio beat.
11. electrocardiogram (ECG) data disposal systems according to claim 10, is characterized in that, described system also comprises:
Information receiver, for receiving according to described representative electrocardio beat feedack;
Feedback processing device, for processing described cluster population according to described feedack.
12. electrocardiogram (ECG) data treating apparatus according to claim 11, is characterized in that, described feedback processing device also comprises:
Signal judgement module, for judging the cluster instruction of whether attaching most importance to of described feedack, if so, notifies reunion generic module, if not, notifies extraction module;
Described reunion generic module is used for adjusting clustering parameter, and the corresponding electrocardiogram (ECG) data of described cluster population is carried out to cluster again, the cluster population of the class that obtains meeting again;
Described extraction module is used for extracting the template that described feedack records, and shows the corresponding result of template of described extraction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310041662.9A CN103970975B (en) | 2013-02-02 | 2013-02-02 | Electrocardiogram (ECG) data processing method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310041662.9A CN103970975B (en) | 2013-02-02 | 2013-02-02 | Electrocardiogram (ECG) data processing method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103970975A true CN103970975A (en) | 2014-08-06 |
| CN103970975B CN103970975B (en) | 2018-07-06 |
Family
ID=51240465
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310041662.9A Active CN103970975B (en) | 2013-02-02 | 2013-02-02 | Electrocardiogram (ECG) data processing method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103970975B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104398254A (en) * | 2014-11-14 | 2015-03-11 | 中国科学院深圳先进技术研究院 | Electrocardiogram analyzing system, electrocardiogram analyzing equipment and electrocardiogram predication model acquisition equipment |
| CN105654046A (en) * | 2015-12-29 | 2016-06-08 | 中国科学院深圳先进技术研究院 | Electrocardiosignal identity identification method and electrocardiosignal identity identification device |
| CN105748066A (en) * | 2016-03-03 | 2016-07-13 | 深圳竹信科技有限公司 | Method and device for extracting feature point of electrocardiosignal waveform |
| CN106618552A (en) * | 2016-12-30 | 2017-05-10 | 清华大学 | Detection model acquisition method and device |
| CN107092789A (en) * | 2017-04-18 | 2017-08-25 | 成都琅瑞医疗技术股份有限公司 | Electrocardiogram (ECG) data analysis tool system and application method |
| CN107622259A (en) * | 2017-10-20 | 2018-01-23 | 山东大学 | A kind of T wave detecting methods, electrocardiogram (ECG) data analysis method and device |
| CN109171706A (en) * | 2018-09-30 | 2019-01-11 | 南京信息工程大学 | The Denoising of ECG Signal and system spread based on classification and matching and fractional order |
| CN112635047A (en) * | 2020-09-22 | 2021-04-09 | 广东工业大学 | Robust electrocardio R peak detection method |
| CN112773382A (en) * | 2021-01-20 | 2021-05-11 | 钛虎机器人科技(上海)有限公司 | Myoelectricity sensing method and system with user self-adaption capability |
-
2013
- 2013-02-02 CN CN201310041662.9A patent/CN103970975B/en active Active
Non-Patent Citations (7)
| Title |
|---|
| SERKAN KIRANYAZ ET AL: "Personalized long-term ECG classification: A systematic approach", 《EXPERT SYSTEMS WITH APPLICATIONS》 * |
| 于天: "一种在线动态心电信号聚类分析策略", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 季虎: "心电信号自动分析关键技术研究", 《中国博士学位论文全文数据库 信息科技辑》 * |
| 曹国超: "动态心电图波形改进分类策略研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 段会龙等: "基于模板匹配和特征识别相结合的心室期前收缩波形分类算法", 《航天医学与医学工程》 * |
| 汤征: "引入异常心电节律分析的心拍分类算法研究", 《中国优秀硕士学位论文全文数据库 医药卫生科技辑》 * |
| 王鸿鹏等: "心电自动分析系统", 《北京生物医学工程》 * |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104398254A (en) * | 2014-11-14 | 2015-03-11 | 中国科学院深圳先进技术研究院 | Electrocardiogram analyzing system, electrocardiogram analyzing equipment and electrocardiogram predication model acquisition equipment |
| CN105654046A (en) * | 2015-12-29 | 2016-06-08 | 中国科学院深圳先进技术研究院 | Electrocardiosignal identity identification method and electrocardiosignal identity identification device |
| CN105654046B (en) * | 2015-12-29 | 2019-01-18 | 中国科学院深圳先进技术研究院 | Electrocardiosignal personal identification method and device |
| CN105748066A (en) * | 2016-03-03 | 2016-07-13 | 深圳竹信科技有限公司 | Method and device for extracting feature point of electrocardiosignal waveform |
| WO2017148452A1 (en) * | 2016-03-03 | 2017-09-08 | 深圳竹信科技有限公司 | Electrocardiography signal waveform feature point extraction method and device |
| CN105748066B (en) * | 2016-03-03 | 2018-03-16 | 深圳竹信科技有限公司 | A kind of extracting method and device of electro-cardiologic signal waveforms characteristic point |
| CN106618552A (en) * | 2016-12-30 | 2017-05-10 | 清华大学 | Detection model acquisition method and device |
| CN106618552B (en) * | 2016-12-30 | 2019-08-09 | 清华大学 | A kind of detection model acquisition methods and device |
| CN107092789A (en) * | 2017-04-18 | 2017-08-25 | 成都琅瑞医疗技术股份有限公司 | Electrocardiogram (ECG) data analysis tool system and application method |
| CN107622259B (en) * | 2017-10-20 | 2019-12-24 | 山东大学 | T wave detection method, electrocardiogram data analysis method and device |
| CN107622259A (en) * | 2017-10-20 | 2018-01-23 | 山东大学 | A kind of T wave detecting methods, electrocardiogram (ECG) data analysis method and device |
| CN109171706A (en) * | 2018-09-30 | 2019-01-11 | 南京信息工程大学 | The Denoising of ECG Signal and system spread based on classification and matching and fractional order |
| CN109171706B (en) * | 2018-09-30 | 2020-12-29 | 南京信息工程大学 | Electrocardiosignal denoising method and system based on classification matching and fractional order diffusion |
| CN112635047A (en) * | 2020-09-22 | 2021-04-09 | 广东工业大学 | Robust electrocardio R peak detection method |
| CN112635047B (en) * | 2020-09-22 | 2021-09-07 | 广东工业大学 | Robust electrocardio R peak detection method |
| CN112773382A (en) * | 2021-01-20 | 2021-05-11 | 钛虎机器人科技(上海)有限公司 | Myoelectricity sensing method and system with user self-adaption capability |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103970975B (en) | 2018-07-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103970975A (en) | Electrocardio data processing method and electrocardio data processing system | |
| CN107714023B (en) | Static electrocardiogram analysis method and device based on artificial intelligence self-learning | |
| CN101065058B (en) | Monitoring physiological activity using partial state space reconstruction | |
| CN103110417B (en) | Automatic electrocardiogram recognition system | |
| US11350868B2 (en) | Electrocardiogram information processing method and electrocardiogram workstation system | |
| US7412282B2 (en) | Algorithms for detecting cardiac arrhythmia and methods and apparatuses utilizing the algorithms | |
| CN109171712A (en) | Auricular fibrillation recognition methods, device, equipment and computer readable storage medium | |
| CN107837082A (en) | Electrocardiogram automatic analysis method and device based on artificial intelligence self study | |
| CN109411042B (en) | Electrocardio information processing method and electrocardio workstation | |
| CN108647565A (en) | A kind of data preprocessing method classified to electrocardiosignal based on deep learning model | |
| CN108511055B (en) | Ventricular premature beat recognition system and method based on classifier fusion and diagnosis rules | |
| CN103690156A (en) | Heart rate acquiring method and electrocardiosignal processing method | |
| WO2009100133A1 (en) | System, method and computer program product for detection of changes in health status and risk of imminent illness | |
| CN107688553B (en) | Method for detecting electrocardiographic waveform characteristics based on wavelet transform and logistic regression algorithm | |
| CN105796096A (en) | Heart rate variability analysis method, heart rate variability analysis system and terminal | |
| CN109009073A (en) | Atrial fibrillation detection device and storage medium | |
| CN108836310A (en) | A kind of method and system for judging automatically user's electrocardio state based on artificial intelligence | |
| CN109222964A (en) | atrial fibrillation detection device and storage medium | |
| CN109044348A (en) | atrial fibrillation detection device and storage medium | |
| CN111297340A (en) | Movement state monitoring method based on combination of GPS and heart rate | |
| CN109431487B (en) | Method for identifying typical data of ventricular heart beat in electrocardiogram dynamic real-time analysis data | |
| CN1180746C (en) | Cardioelectric spectrometer | |
| CN106974644A (en) | A kind of electrocardiogram R wave detection method and system | |
| Zhang et al. | An improved real-time R-wave detection efficient algorithm in exercise ECG signal analysis | |
| CN109044338A (en) | atrial fibrillation detection device and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |