CN104834921A - Electrocardio normality/abnormality big-data processing method and device - Google Patents
Electrocardio normality/abnormality big-data processing method and device Download PDFInfo
- Publication number
- CN104834921A CN104834921A CN201510268861.2A CN201510268861A CN104834921A CN 104834921 A CN104834921 A CN 104834921A CN 201510268861 A CN201510268861 A CN 201510268861A CN 104834921 A CN104834921 A CN 104834921A
- Authority
- CN
- China
- Prior art keywords
- data
- ecg
- electrocardiogram
- wave
- sorted
- 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
- G06F2218/16—Classification; Matching by matching signal segments
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/03—Recognition of patterns in medical or anatomical images
Abstract
The invention discloses an electrocardio normality/abnormality big-data processing method and device, and the method employs a normal electrocardio database, and comprises the following steps: S1, partitioning to-be-classified electrocardio data according to cardiac beat, and then carrying out the normalization processing of length and amplitude, thereby forming a plurality of pieces of beat-wave-shaped data; S2, extracting the index data of to-be-classified electrocardio data; S3, determining a confidence interval according to the index data stored in the database, comparing the index data of the extracted to-be-classified electrocardio data with the confidence interval, and outputting a comparison result; S4, calculating the similarity of the plurality of pieces of beat-wave-shaped data obtained through the partitioning of the to-be-classified electrocardio data with waveform data, corresponding to cardiac beat, in the electrocardio data stored in the database, comparing the similarity with a similarity threshold value, and outputting the comparison results. The device comprises the normal electrocardio database, and a plurality of modules which are used for achieving the above steps. The device can achieve the reliable classification and screening of the electrocardiogram which is not diagnosed, and avoids false negative.
Description
Technical field
The present invention relates to pattern-recognition, large data analysis, medical signals process field, specifically a kind of electrocardio just/abnormal large data processing method and device.
Background technology
Cardiogram is a clinical conventional inspection, by the situation of each cycle electrical activity of record heart, can assisted diagnosis arrhythmia cordis, impatiently ischemic, miocardial infarction, atrioventricular hypertrophy, block, premature beat etc. extremely, also can be used for judging that electrolyte and medicine are on the impact etc. of heart.
At present, popularizing of electrocardio examination, is limited to a great extent and can understands Electrocardiographic Cardiologists quantity.So many research institutions are devoted to the research and development of electrocardio auto-check system.Electrocardio automatic classification is realized both at home and abroad for computing machine, have employed a variety of research method, as artificial neural network, fuzzy clustering, logic discrimination tree, template matches ..., when classifying to electrocardio, what have is divided into multiple kind according to different diseases by abnormal electrocardiogram; What have classifies to it for certain specific exceptions (as premature ventricualr contraction).Not yet there is a kind of more ripe method, whether supporting electrocardio examination by identifying electrocardiographic abnormality.
Summary of the invention
The object of this invention is to provide a kind of electrocardio just/abnormal large data processing method and device, make it possible to reliably sort out examination by computing machine to ND cardiogram (electrocardiogram (ECG) data to be sorted), avoid false negative (abnormal electrocardiogram is judged as normal electrocardio), thus make doctor only need being judged as that abnormal carrying out is diagnosed again, reduce the workload of doctor.
Concrete technical scheme of the present invention is:
A kind of electrocardio just/abnormal large data processing method, this disposal route comprises normal ecg database, this database purchase has normal electrocardiogram (ECG) data as much as possible, and the normal electrocardiogram (ECG) data of every bar comprises achievement data and claps the some bat Wave datas obtained electrocardiogram (ECG) data segmentation by the heart; This disposal route comprises the following steps:
S1, by the heart clap electrocardiogram (ECG) data to be sorted is split, then respectively normalized is done to length and amplitude, forms some bat Wave datas;
S2, extract the achievement data of described electrocardiogram (ECG) data to be sorted;
S3, achievement data determination fiducial interval according to described database purchase, and the achievement data of the electrocardiogram (ECG) data to be sorted extracted is compared with described fiducial interval, export comparative result; And
S4, calculate the similarity of the Wave data that some bat Wave datas of being divided into from electrocardiogram (ECG) data to be sorted are clapped to the corresponding heart the electrocardiogram (ECG) data of described database purchase, compare output comparative result with similarity threshold;
Described achievement data comprises QRS wave band length, PR interval, at least one interim between QT interval and RR.
Above-mentioned electrocardio just/abnormal large data processing method in, preferably, in described normal ecg database, the corresponding each heart of every bar electrocardiogram (ECG) data is clapped and is comprised multistage Wave data, described multistage Wave data length is equal, the overwhelming majority is overlapping, and the front and back that the center of described multistage Wave data lays respectively at waveform peak place and this waveform peak only differ some data points each other; Described step S4 comprises the following steps:
S41, clap the described multistage Wave data clapped to the corresponding heart in the electric data of uniting as one of described database purchase respectively of Wave data by be divided into from described electrocardiogram (ECG) data to be sorted one and calculate, obtain the multiple similarities relative to described multistage Wave data;
S42, from the multiple similarities relative to described multistage Wave data obtained, choose minimum value, clap the similarity of the Wave data that Wave data is clapped to the corresponding heart in electric data of uniting as one described in described database purchase as be divided into from described electrocardiogram (ECG) data to be sorted described one;
S43, circulation perform described step S41 and S42, calculate be divided into from described electrocardiogram (ECG) data to be sorted described one clap the similarity of Wave data that Wave data claps to the corresponding heart other electrocardiogram (ECG) data of described database purchase; And
S44, circulation perform described step S41, S42, S43, obtain the similarity that other that be divided into from described electrocardiogram (ECG) data to be sorted claps the Wave data that Wave data is clapped to the corresponding heart the electrocardiogram (ECG) data of described database purchase.
Above-mentioned electrocardio just/abnormal large data processing method in, preferably, electrocardiogram (ECG) data to be sorted is carried out pretreated step before being also included in segmentation and index extraction by this disposal route.
Above-mentioned electrocardio just/abnormal large data processing method in, preferably, the comparative result that described step S4 exports comprises: normal electrocardiogram (ECG) data or abnormal electrocardiogram data, when the comparative result exported is abnormal electrocardiogram data, the comparative result exported also comprises the cycle that there is exception, described exist the abnormal cycle and refer in some bat Wave datas of described electrocardiogram (ECG) data to be sorted, and the heart that the similarity of the Wave data clapped to the corresponding heart in each bar electrocardiogram (ECG) data of described database purchase is all greater than the Wave data of described similarity threshold corresponding claps the cycle.
Above-mentioned electrocardio just/abnormal large data processing method in, preferably, in described step S4, Similarity Measure comprises: the corresponding point of to be compared two sections of Wave datas are asked difference one by one; And summing value after each difference is taken absolute value, using this and the similarity of value as these two sections of Wave datas.
A kind of electrocardio just/abnormal large data processing equipment, this treating apparatus comprises:
Normal ecg database, this database purchase has normal electrocardiogram (ECG) data as much as possible, and the normal electrocardiogram (ECG) data of every bar comprises achievement data and claps the some bat Wave datas obtained electrocardiogram (ECG) data segmentation by the heart;
Segmentation module, splitting electrocardiogram (ECG) data to be sorted for clapping by the heart, then doing normalized to length and amplitude respectively, forming some bat Wave datas;
Index extraction module, for extracting the achievement data of described electrocardiogram (ECG) data to be sorted;
Targets match module, for the achievement data determination fiducial interval according to described database purchase, and compares the achievement data of the electrocardiogram (ECG) data to be sorted extracted with described fiducial interval, and exports comparative result; And
Waveform Matching module, for calculating the similarity of the Wave data that some bat Wave datas of being divided into from electrocardiogram (ECG) data to be sorted are clapped to the corresponding heart the electrocardiogram (ECG) data of described database purchase, compares output comparative result with similarity threshold;
Described achievement data comprises QRS wave band length, PR interval, at least one interim between QT interval and RR.
Above-mentioned electrocardio just/abnormal large data processing equipment in, preferably, in described normal ecg database, the corresponding each heart of every bar electrocardiogram (ECG) data is clapped and is comprised multistage Wave data, described multistage Wave data length is equal, the overwhelming majority is overlapping, and the front and back that the center of described multistage Wave data lays respectively at waveform peak place and this waveform peak only differ some data points each other; Described Waveform Matching module comprises:
First module, the described multistage Wave data clapped to the corresponding heart in the electric data of uniting as one of described database purchase respectively for clapping Wave data by be divided into from described electrocardiogram (ECG) data to be sorted one calculates, and obtains the multiple similarities relative to described multistage Wave data;
Second module, for choosing minimum value from the multiple similarities relative to described multistage Wave data obtained, clap the similarity of the Wave data that Wave data is clapped to the corresponding heart in electric data of uniting as one described in described database purchase as be divided into from described electrocardiogram (ECG) data to be sorted described one;
3rd module, for the first module described in recursive call and the second module, calculate be divided into from described electrocardiogram (ECG) data to be sorted described one clap the similarity of Wave data that Wave data claps to the corresponding heart other electrocardiogram (ECG) data of described database purchase; And
Four module, for the first module, the second module and the 3rd module described in recursive call, obtains the similarity that other that be divided into from described electrocardiogram (ECG) data to be sorted claps the Wave data that Wave data is clapped to the corresponding heart the electrocardiogram (ECG) data of described database purchase.
Above-mentioned electrocardio just/abnormal large data processing equipment in, preferably, this treating apparatus also comprises pretreatment module, for electrocardiogram (ECG) data to be sorted being carried out pre-service before segmentation and index extraction.
Above-mentioned electrocardio just/abnormal large data processing equipment in, preferably, the comparative result that described Waveform Matching module exports comprises: normal electrocardiogram (ECG) data or abnormal electrocardiogram data, when the comparative result exported is abnormal electrocardiogram data, the comparative result exported also comprises the cycle that there is exception, described exist the abnormal cycle and refer in some bat Wave datas of described electrocardiogram (ECG) data to be sorted, and the heart that the similarity of the Wave data clapped to the corresponding heart in each bar electrocardiogram (ECG) data of described database purchase is all greater than the Wave data of described similarity threshold corresponding claps the cycle.
Above-mentioned electrocardio just/abnormal large data processing equipment in, preferably, in described Waveform Matching module, Similarity Measure comprises: the corresponding point of to be compared two sections of Wave datas are asked difference one by one; And summing value after each difference is taken absolute value, using this and the similarity of value as these two sections of Wave datas.
The present invention is by the law mining of normal ecg database and analysis, the matching algorithm of feature based and waveform realizes electrocardiogram (ECG) data classification, can reliably identify normal electrocardiogram (ECG) data, avoid occurring false negative (abnormal electrocardiogram is judged as normal electrocardio), therefore, can carry out electrocardio examination by assist physician, doctor only needs being judged as that abnormal carrying out is diagnosed again, thus can greatly reduce the workload of doctor.
Accompanying drawing explanation
Fig. 1 be electrocardio of the present invention just/process flow diagram of abnormal large some embodiments of data processing method;
Fig. 2-Fig. 6 is the match condition of five sections of waveforms that the bat waveform be divided into from electrocardiogram (ECG) data to be sorted is clapped to the corresponding heart the electric data of uniting as one of database purchase.
Embodiment
Below in conjunction with drawings and Examples, the present invention is further described.These more detailed descriptions are intended to help and understand the present invention, and should not be used to limit the present invention.According to content disclosed by the invention, it will be understood by those skilled in the art that and some or all these specific detail can not be needed to implement the present invention.And in other cases, in order to avoid innovation and creation being desalinated, do not describe well-known operating process in detail.
As shown in Figure 1, the electrocardio of some embodiments just/abnormal large data processing method comprises normal ecg database, this database purchase has normal electrocardiogram (ECG) data as much as possible, and the normal electrocardiogram (ECG) data of every bar comprises achievement data and claps the some bat Wave datas obtained electrocardiogram (ECG) data segmentation by the heart; Wherein, described achievement data comprises QRS wave band length, PR interval, QT interval and RR interval.
As a rule, by the collection of up to ten thousand normal electrocardiogram (ECG) datas and arrangement, normal ecg database can be set up, for the summary of normal electrocardiogram (ECG) data rule and analysis.The concrete process of establishing of normal ecg database is as follows:
Step S100, by peak-value detection method each cycle of normal for every bar electrocardiogram (ECG) data is intercepted out (namely normal electrocardiogram (ECG) data press the heart clap segmentation).Suppose in data, the data point that a certain cycle intercepts is kth point ~ kth+N point, then kth point ~ kth+N point, kth-2 points ~ kth+N-2 point, kth-1 point ~ kth+N-1, kth+1 point ~ kth+N+1 point, kth+2 points ~ kth+N+2 are put five segment datas respectively stored in database, stored in doing length normalization method by methods such as interpolation by every section that intercepts before, and normalization is done to the amplitude of different segment data.Therefore, the corresponding each heart of every bar electrocardiogram (ECG) data is clapped and is comprised multistage (in this example being five sections) Wave data, described multistage Wave data length is equal, the overwhelming majority is overlapping, and the front and back that the center of described multistage Wave data lays respectively at waveform peak place and this waveform peak only differ some data points each other.
Step S200, by the index such as QRS wave band length, PR interval, QT interval, RR interval of normal for every bar electrocardiogram (ECG) data stored in database.
The electrocardio of some embodiments just/abnormal large data processing method comprises the following steps:
Step S1, by the heart clap electrocardiogram (ECG) data to be sorted is split, then respectively normalized is done to length and amplitude, forms some bat Wave datas.Particularly, according to the result of QRS wave group identification, can split often clapping waveform.Can the methods such as dual threshold be adopted during segmentation, determine each cycle segmentation starting point and terminal.Suppose that bat Wave data (a certain cycle data) be divided into from electrocardiogram (ECG) data to be sorted is a={a
1, a
2..., a
n, in order to make it consistent with every section of Wave data length in database, it is carried out to the operations such as interpolation, the data after process are
Step S2, extract the achievement data of described electrocardiogram (ECG) data to be sorted.Described achievement data comprises QRS wave band length, PR interval, QT interval and RR interval.
Step S3, achievement data determination fiducial interval according to described database purchase, and the achievement data of the electrocardiogram (ECG) data to be sorted extracted is compared with described fiducial interval, export comparative result.That is, first by targets match, preliminary judgement is made to the whether normal of described electrocardiogram (ECG) data to be sorted.Particularly, according to indication range, average, the variance such as QRS wave band length, PR interval, QT interval, RR interval of electrocardiogram (ECG) data normal in database, fiducial interval can be determined, for making preliminary judgement to the whether normal of electrocardiogram (ECG) data to be detected.
Step S4, calculate the similarity of the Wave data that some bat Wave datas of being divided into from electrocardiogram (ECG) data to be sorted are clapped to the corresponding heart the electrocardiogram (ECG) data of described database purchase, compare output comparative result with similarity threshold.I.e. Waveform Matching step.
Mainly through not diagnosing the waveform of cardiogram (electrocardiogram (ECG) data to be sorted) and known normal cardiac electrical similarity in Waveform Matching step, whether it is normally judged.
Consider that similarity analysis is with closely related by the relative position of two waveforms that compares, so when carrying out Similarity Measure, when calculating two waveform peak alignment and multiple similarities of obtaining when moving some data points of peak value relative position, therefrom get minimum value as the final similarity of this two waveforms simultaneously.Therefore, described step S4 preferably includes following steps:
S41, clap the described multistage Wave data clapped to the corresponding heart in the electric data of uniting as one of described database purchase respectively of Wave data by be divided into from described electrocardiogram (ECG) data to be sorted one and calculate, obtain the multiple similarities relative to described multistage Wave data.Particularly, following formula can be expressed as
Wherein,
represent the bat Wave data be divided into from described electrocardiogram (ECG) data to be sorted, b
i, j, k, wrepresent the multistage Wave data in a jth cycle (heart bat) in i-th electrocardiogram (ECG) data stored in database, p=1,2, ..., N, k=1,2, ..., N represents the length of Wave data, w=1,2, ..., the hop count of 5 expression multistage Wave datas, concrete corresponding each cycle (setting starting point as kth point) is stored in 5 segment datas (k-2 ~ k+N-2 point, k-1 point ~ k+N-1, k ~ k+N point, k+1 ~ k+N+1 point, k+2 point ~ k+N+2 point) in database.
Therefore in the present embodiment, Similarity Measure comprises: the corresponding point of to be compared two sections of Wave datas are asked difference one by one; And summing value after each difference is taken absolute value, using this and the similarity of value as these two sections of Wave datas.But the present invention is not limited to this, such as, averages after can also taking absolute value to each difference, using the similarity of this average as these two sections of Wave datas, also can adopt other similar approach.
The match condition of five sections of waveforms 2,3,4,5,6 that the bat waveform 1 be divided into from electrocardiogram (ECG) data to be sorted is clapped to the corresponding heart the electric data of uniting as one of database purchase has been shown in Fig. 2-Fig. 6.
S42, from the multiple similarities relative to described multistage Wave data obtained, choose minimum value, clap the similarity of the Wave data that Wave data is clapped to the corresponding heart in electric data of uniting as one described in described database purchase as be divided into from described electrocardiogram (ECG) data to be sorted described one.Following formula can be expressed as
S43, circulation perform described step S41 and S42, calculate be divided into from described electrocardiogram (ECG) data to be sorted described one clap the similarity of Wave data that Wave data claps to the corresponding heart other electrocardiogram (ECG) data of described database purchase.That is, in above-mentioned formula, i is traveled through, j.By the similarity Δ calculated
i, jcompare with similarity threshold θ, when there is Δ
i, jduring > θ, judge this section of waveform
be a normal cardiac electrical cycle, wherein similarity threshold θ can set based on experience value.
S44, circulation perform described step S41, S42, S43, obtain the similarity that other that be divided into from described electrocardiogram (ECG) data to be sorted claps the Wave data that Wave data is clapped to the corresponding heart the electrocardiogram (ECG) data of described database purchase.If the similarity of all sections is all less than similarity threshold θ, then, Waveform Matching Output rusults for this electrocardiogram (ECG) data be normal electrocardiogram (ECG) data, otherwise to export this electrocardiogram (ECG) data be abnormal electrocardiogram data, and export and there is the abnormal cycle.
When These parameters coupling and Waveform Matching result are all normal, judge that this electrocardiogram (ECG) data to be sorted is normal, otherwise be abnormal.
Further, in certain embodiments, before being also included in segmentation and index extraction, electrocardiogram (ECG) data to be sorted is carried out pretreated step S1 '.These pre-service can comprise: by basic electrocardiographicdata data Fast Classification, as when heart rate is abnormal, directly judge that electrocardiogram (ECG) data is abnormal; For the normal cardiogram of basic electrocardiographicdata data, if baseline wander or noise are seriously, by modes such as filtering, denoising is carried out to it.
Particularly point out, in the present invention, the sequence number of step is only for convenience of description, do not represent specific ordinal relation, such as, state step S1 in method, S2, S3, S4 not must by this execution that puts in order, the order of step S1 and S2 can be exchanged, and step S1 also can be placed on after step S3, etc.
With more above-mentioned embodiment electrocardios just/electrocardio corresponding to abnormal large data processing method just/abnormal large data processing equipment, comprising:
Normal ecg database, this database purchase has normal electrocardiogram (ECG) data as much as possible, and the normal electrocardiogram (ECG) data of every bar comprises achievement data and claps the some bat Wave datas obtained electrocardiogram (ECG) data segmentation by the heart;
Segmentation module, splitting electrocardiogram (ECG) data to be sorted for clapping by the heart, then doing normalized to length and amplitude respectively, forming some bat Wave datas;
Index extraction module, for extracting the achievement data of described electrocardiogram (ECG) data to be sorted;
Targets match module, for the achievement data determination fiducial interval according to described database purchase, and compares the achievement data of the electrocardiogram (ECG) data to be sorted extracted with described fiducial interval, and exports comparative result; And
Waveform Matching module, for calculating the similarity of the Wave data that some bat Wave datas of being divided into from electrocardiogram (ECG) data to be sorted are clapped to the corresponding heart the electrocardiogram (ECG) data of described database purchase, compares output comparative result with similarity threshold;
Described achievement data comprises QRS wave band length, PR interval, at least one interim between QT interval and RR.
In described normal ecg database, the corresponding each heart of every bar electrocardiogram (ECG) data is clapped and is comprised multistage Wave data, described multistage Wave data length is equal, the overwhelming majority is overlapping, and the front and back that the center of described multistage Wave data lays respectively at waveform peak place and this waveform peak only differ some data points each other; Described Waveform Matching module comprises:
First module, the described multistage Wave data clapped to the corresponding heart in the electric data of uniting as one of described database purchase respectively for clapping Wave data by be divided into from described electrocardiogram (ECG) data to be sorted one calculates, and obtains the multiple similarities relative to described multistage Wave data;
Second module, for choosing minimum value from the multiple similarities relative to described multistage Wave data obtained, clap the similarity of the Wave data that Wave data is clapped to the corresponding heart in electric data of uniting as one described in described database purchase as be divided into from described electrocardiogram (ECG) data to be sorted described one;
3rd module, for the first module described in recursive call and the second module, calculate be divided into from described electrocardiogram (ECG) data to be sorted described one clap the similarity of Wave data that Wave data claps to the corresponding heart other electrocardiogram (ECG) data of described database purchase; And
Four module, for the first module, the second module and the 3rd module described in recursive call, obtains the similarity that other that be divided into from described electrocardiogram (ECG) data to be sorted claps the Wave data that Wave data is clapped to the corresponding heart the electrocardiogram (ECG) data of described database purchase.
This treating apparatus also comprises pretreatment module, for electrocardiogram (ECG) data to be sorted being carried out pre-service before segmentation and index extraction.
The comparative result that described Waveform Matching module exports comprises: normal electrocardiogram (ECG) data or abnormal electrocardiogram data, when the comparative result exported is abnormal electrocardiogram data, the comparative result exported also comprises the cycle that there is exception, described exist the abnormal cycle and refer in some bat Wave datas of described electrocardiogram (ECG) data to be sorted, and the heart that the similarity of the Wave data clapped to the corresponding heart in each bar electrocardiogram (ECG) data of described database purchase is all greater than the Wave data of described similarity threshold corresponding claps the cycle.
In described Waveform Matching module, Similarity Measure comprises: the corresponding point of to be compared two sections of Wave datas are asked difference one by one; And summing value after each difference is taken absolute value, using this and the similarity of value as these two sections of Wave datas.
In the above-described embodiments, by the law mining of normal ecg database and analysis, the matching algorithm of feature based and waveform realizes electrocardiogram (ECG) data classification, can reliably identify normal electrocardiogram (ECG) data, avoid false-negative appearance.Therefore, can diagnose a large amount of electrocardiogram (ECG) data by assist physician, namely before doctor diagnoses a large amount of electrocardiogram (ECG) data, first use computing machine to carry out automatic examination by said method or device, the abnormal electrocardiogram data that doctor only need go out examination are diagnosed.
Claims (10)
1. an electrocardio just/abnormal large data processing method, it is characterized in that, this disposal route comprises normal ecg database, this database purchase has normal electrocardiogram (ECG) data as much as possible, and the normal electrocardiogram (ECG) data of every bar comprises achievement data and claps the some bat Wave datas obtained electrocardiogram (ECG) data segmentation by the heart; This disposal route comprises the following steps:
S1, by the heart clap electrocardiogram (ECG) data to be sorted is split, then respectively normalized is done to length and amplitude, forms some bat Wave datas;
S2, extract the achievement data of described electrocardiogram (ECG) data to be sorted;
S3, achievement data determination fiducial interval according to described database purchase, and the achievement data of the electrocardiogram (ECG) data to be sorted extracted is compared with described fiducial interval, export comparative result; And
S4, calculate the similarity of the Wave data that some bat Wave datas of being divided into from electrocardiogram (ECG) data to be sorted are clapped to the corresponding heart the electrocardiogram (ECG) data of described database purchase, compare output comparative result with similarity threshold;
Described achievement data comprises QRS wave band length, PR interval, at least one interim between QT interval and RR.
2. electrocardio according to claim 1 just/abnormal large data processing method, it is characterized in that,
In described normal ecg database, the corresponding each heart of every bar electrocardiogram (ECG) data is clapped and is comprised multistage Wave data, described multistage Wave data length is equal, the overwhelming majority is overlapping, and the front and back that the center of described multistage Wave data lays respectively at waveform peak place and this waveform peak only differ some data points each other;
Described step S4 comprises the following steps:
S41, clap the described multistage Wave data clapped to the corresponding heart in the electric data of uniting as one of described database purchase respectively of Wave data by be divided into from described electrocardiogram (ECG) data to be sorted one and calculate, obtain the multiple similarities relative to described multistage Wave data;
S42, from the multiple similarities relative to described multistage Wave data obtained, choose minimum value, clap the similarity of the Wave data that Wave data is clapped to the corresponding heart in electric data of uniting as one described in described database purchase as be divided into from described electrocardiogram (ECG) data to be sorted described one;
S43, circulation perform described step S41 and S42, calculate be divided into from described electrocardiogram (ECG) data to be sorted described one clap the similarity of Wave data that Wave data claps to the corresponding heart other electrocardiogram (ECG) data of described database purchase; And
S44, circulation perform described step S41, S42, S43, obtain the similarity that other that be divided into from described electrocardiogram (ECG) data to be sorted claps the Wave data that Wave data is clapped to the corresponding heart the electrocardiogram (ECG) data of described database purchase.
3. electrocardio according to claim 1 just/abnormal large data processing method, it is characterized in that, electrocardiogram (ECG) data to be sorted is carried out pretreated step before being also included in segmentation and index extraction by this disposal route.
4. electrocardio according to claim 1 just/abnormal large data processing method, it is characterized in that, the comparative result that described step S4 exports comprises: normal electrocardiogram (ECG) data or abnormal electrocardiogram data, when the comparative result exported is abnormal electrocardiogram data, the comparative result exported also comprises the cycle that there is exception, the described cycle that there is exception refers in some bat Wave datas of described electrocardiogram (ECG) data to be sorted, the heart that the similarity of the Wave data clapped to the corresponding heart in each bar electrocardiogram (ECG) data of described database purchase is all greater than the Wave data of described similarity threshold corresponding claps the cycle.
5. electrocardio according to claim 1 just/abnormal large data processing method, it is characterized in that, in described step S4, Similarity Measure comprises: the corresponding point of to be compared two sections of Wave datas are asked difference one by one; And summing value after each difference is taken absolute value, using this and the similarity of value as these two sections of Wave datas.
6. electrocardio just/an abnormal large data processing equipment, it is characterized in that, this treating apparatus comprises:
Normal ecg database, this database purchase has normal electrocardiogram (ECG) data as much as possible, and the normal electrocardiogram (ECG) data of every bar comprises achievement data and claps the some bat Wave datas obtained electrocardiogram (ECG) data segmentation by the heart;
Segmentation module, splitting electrocardiogram (ECG) data to be sorted for clapping by the heart, then doing normalized to length and amplitude respectively, forming some bat Wave datas;
Index extraction module, for extracting the achievement data of described electrocardiogram (ECG) data to be sorted;
Targets match module, for the achievement data determination fiducial interval according to described database purchase, and compares the achievement data of the electrocardiogram (ECG) data to be sorted extracted with described fiducial interval, and exports comparative result; And
Waveform Matching module, for calculating the similarity of the Wave data that some bat Wave datas of being divided into from electrocardiogram (ECG) data to be sorted are clapped to the corresponding heart the electrocardiogram (ECG) data of described database purchase, compares output comparative result with similarity threshold;
Described achievement data comprises QRS wave band length, PR interval, at least one interim between QT interval and RR.
7. electrocardio according to claim 6 just/abnormal large data processing equipment, it is characterized in that,
In described normal ecg database, the corresponding each heart of every bar electrocardiogram (ECG) data is clapped and is comprised multistage Wave data, described multistage Wave data length is equal, the overwhelming majority is overlapping, and the front and back that the center of described multistage Wave data lays respectively at waveform peak place and this waveform peak only differ some data points each other;
Described Waveform Matching module comprises:
First module, the described multistage Wave data clapped to the corresponding heart in the electric data of uniting as one of described database purchase respectively for clapping Wave data by be divided into from described electrocardiogram (ECG) data to be sorted one calculates, and obtains the multiple similarities relative to described multistage Wave data;
Second module, for choosing minimum value from the multiple similarities relative to described multistage Wave data obtained, clap the similarity of the Wave data that Wave data is clapped to the corresponding heart in electric data of uniting as one described in described database purchase as be divided into from described electrocardiogram (ECG) data to be sorted described one;
3rd module, for the first module described in recursive call and the second module, calculate be divided into from described electrocardiogram (ECG) data to be sorted described one clap the similarity of Wave data that Wave data claps to the corresponding heart other electrocardiogram (ECG) data of described database purchase; And
Four module, for the first module, the second module and the 3rd module described in recursive call, obtains the similarity that other that be divided into from described electrocardiogram (ECG) data to be sorted claps the Wave data that Wave data is clapped to the corresponding heart the electrocardiogram (ECG) data of described database purchase.
8. electrocardio according to claim 6 just/abnormal large data processing equipment, it is characterized in that, this treating apparatus also comprises pretreatment module, for electrocardiogram (ECG) data to be sorted being carried out pre-service before segmentation and index extraction.
9. electrocardio according to claim 6 just/abnormal large data processing equipment, it is characterized in that, the comparative result that described Waveform Matching module exports comprises: normal electrocardiogram (ECG) data or abnormal electrocardiogram data, when the comparative result exported is abnormal electrocardiogram data, the comparative result exported also comprises the cycle that there is exception, the described cycle that there is exception refers in some bat Wave datas of described electrocardiogram (ECG) data to be sorted, the heart that the similarity of the Wave data clapped to the corresponding heart in each bar electrocardiogram (ECG) data of described database purchase is all greater than the Wave data of described similarity threshold corresponding claps the cycle.
10. electrocardio according to claim 6 just/abnormal large data processing equipment, it is characterized in that, in described Waveform Matching module, Similarity Measure comprises: the corresponding point of to be compared two sections of Wave datas are asked difference one by one; And summing value after each difference is taken absolute value, using this and the similarity of value as these two sections of Wave datas.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510268861.2A CN104834921B (en) | 2015-05-22 | 2015-05-22 | Electrocardio just/abnormal big data processing method and processing device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510268861.2A CN104834921B (en) | 2015-05-22 | 2015-05-22 | Electrocardio just/abnormal big data processing method and processing device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104834921A true CN104834921A (en) | 2015-08-12 |
CN104834921B CN104834921B (en) | 2018-01-16 |
Family
ID=53812799
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510268861.2A Expired - Fee Related CN104834921B (en) | 2015-05-22 | 2015-05-22 | Electrocardio just/abnormal big data processing method and processing device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104834921B (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108090509A (en) * | 2017-12-13 | 2018-05-29 | 四川大学 | A kind of adaptive electrocardiogram sorting technique of data length |
CN108227930A (en) * | 2018-01-18 | 2018-06-29 | 四川斐讯信息技术有限公司 | The gestural control method and wearable device of a kind of wearable device |
WO2018166136A1 (en) * | 2017-03-14 | 2018-09-20 | 华为技术有限公司 | Method and apparatus for processing waveform signal |
CN108903956A (en) * | 2018-05-22 | 2018-11-30 | 上海掌门科技有限公司 | It detects a lie device and method |
CN109754877A (en) * | 2018-01-23 | 2019-05-14 | 上海移视网络科技有限公司 | A kind of 12 lead standard cardioelectric figure acute myocardial infarction AMI intelligent distinguishing system based on artificial intelligence |
WO2020031105A3 (en) * | 2018-08-07 | 2020-06-11 | Goldtech Sino Ltd | Noninvasive systems and methods for continuous hemodynamic monitoring |
CN113349791A (en) * | 2021-05-31 | 2021-09-07 | 平安科技(深圳)有限公司 | Abnormal electrocardiosignal detection method, device, equipment and medium |
CN113576422A (en) * | 2021-10-08 | 2021-11-02 | 隽智生物医学研究实验室(佛山)有限公司 | Method, system and equipment for simplifying processing and quantizing physiological signal waveform |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030073917A1 (en) * | 2001-10-12 | 2003-04-17 | Neuropace, Inc. | Patient-specific parameter selection for neurological event detection |
CN103110417A (en) * | 2013-02-28 | 2013-05-22 | 华东师范大学 | Automatic electrocardiogram recognition system |
CN104102915A (en) * | 2014-07-01 | 2014-10-15 | 清华大学深圳研究生院 | Multiple-template matching identity recognition method based on ECG (Electrocardiogram) under electrocardiogram abnormality state |
CN104473633A (en) * | 2014-12-31 | 2015-04-01 | 广州视源电子科技股份有限公司 | Judging method and device of abnormal electrocardio data |
CN104523265A (en) * | 2014-12-31 | 2015-04-22 | 广州视源电子科技股份有限公司 | Judging method and device for abnormal electrocardio data |
-
2015
- 2015-05-22 CN CN201510268861.2A patent/CN104834921B/en not_active Expired - Fee Related
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030073917A1 (en) * | 2001-10-12 | 2003-04-17 | Neuropace, Inc. | Patient-specific parameter selection for neurological event detection |
CN103110417A (en) * | 2013-02-28 | 2013-05-22 | 华东师范大学 | Automatic electrocardiogram recognition system |
CN104102915A (en) * | 2014-07-01 | 2014-10-15 | 清华大学深圳研究生院 | Multiple-template matching identity recognition method based on ECG (Electrocardiogram) under electrocardiogram abnormality state |
CN104473633A (en) * | 2014-12-31 | 2015-04-01 | 广州视源电子科技股份有限公司 | Judging method and device of abnormal electrocardio data |
CN104523265A (en) * | 2014-12-31 | 2015-04-22 | 广州视源电子科技股份有限公司 | Judging method and device for abnormal electrocardio data |
Non-Patent Citations (1)
Title |
---|
聂桥桥: "智能处理在ECG检测中的应用研究", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11321561B2 (en) | 2017-03-14 | 2022-05-03 | Huawei Technologies Co., Ltd. | Electrocardiogram waveform signal processing method and apparatus |
WO2018166136A1 (en) * | 2017-03-14 | 2018-09-20 | 华为技术有限公司 | Method and apparatus for processing waveform signal |
CN108701219A (en) * | 2017-03-14 | 2018-10-23 | 华为技术有限公司 | The method and device of waveform signal processing |
CN108090509A (en) * | 2017-12-13 | 2018-05-29 | 四川大学 | A kind of adaptive electrocardiogram sorting technique of data length |
CN108090509B (en) * | 2017-12-13 | 2021-10-08 | 四川大学 | Data length self-adaptive electrocardiogram classification method |
CN108227930A (en) * | 2018-01-18 | 2018-06-29 | 四川斐讯信息技术有限公司 | The gestural control method and wearable device of a kind of wearable device |
CN109754877A (en) * | 2018-01-23 | 2019-05-14 | 上海移视网络科技有限公司 | A kind of 12 lead standard cardioelectric figure acute myocardial infarction AMI intelligent distinguishing system based on artificial intelligence |
CN109754877B (en) * | 2018-01-23 | 2023-12-22 | 上海移视网络科技有限公司 | Twelve-lead standard electrocardiogram acute myocardial infarction intelligent discrimination system based on artificial intelligence |
CN108903956A (en) * | 2018-05-22 | 2018-11-30 | 上海掌门科技有限公司 | It detects a lie device and method |
WO2020031105A3 (en) * | 2018-08-07 | 2020-06-11 | Goldtech Sino Ltd | Noninvasive systems and methods for continuous hemodynamic monitoring |
CN113349791A (en) * | 2021-05-31 | 2021-09-07 | 平安科技(深圳)有限公司 | Abnormal electrocardiosignal detection method, device, equipment and medium |
CN113576422B (en) * | 2021-10-08 | 2021-12-24 | 隽智生物医学研究实验室(佛山)有限公司 | Method, system and equipment for simplifying processing and quantizing physiological signal waveform |
CN113576422A (en) * | 2021-10-08 | 2021-11-02 | 隽智生物医学研究实验室(佛山)有限公司 | Method, system and equipment for simplifying processing and quantizing physiological signal waveform |
Also Published As
Publication number | Publication date |
---|---|
CN104834921B (en) | 2018-01-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104834921A (en) | Electrocardio normality/abnormality big-data processing method and device | |
CN103110417B (en) | Automatic electrocardiogram recognition system | |
US11134880B2 (en) | Automatic method to delineate or categorize an electrocardiogram | |
Andreotti et al. | Comparing feature-based classifiers and convolutional neural networks to detect arrhythmia from short segments of ECG | |
Yang et al. | Arrhythmia recognition and classification using combined parametric and visual pattern features of ECG morphology | |
CN102835954B (en) | Method for generating heart beating waveform template and treating ventricular premature beat detection and device | |
US7769434B2 (en) | Method of physiological data analysis and measurement quality check using principal component analysis | |
Li et al. | Inter-patient arrhythmia classification with improved deep residual convolutional neural network | |
CN109171712A (en) | Auricular fibrillation recognition methods, device, equipment and computer readable storage medium | |
CN108511055B (en) | Ventricular premature beat recognition system and method based on classifier fusion and diagnosis rules | |
US11331034B2 (en) | Automatic method to delineate or categorize an electrocardiogram | |
Kaya et al. | Feature selection using genetic algorithms for premature ventricular contraction classification | |
CN110522442B (en) | Multi-lead electrocardiographic abnormality detection device, electronic apparatus, and storage medium | |
CN109276242A (en) | The method and apparatus of electrocardiosignal type identification | |
CN111631705A (en) | Electrocardio abnormality detection method, model training method, device, equipment and medium | |
CN112528783A (en) | Electrocardiogram heartbeat data clustering method and device, electronic equipment and medium | |
CN112022141B (en) | Electrocardiosignal class detection method, electrocardiosignal class detection device and storage medium | |
Sharma et al. | Novel T-wave detection technique with minimal processing and RR-interval based enhanced efficiency | |
CN115486855A (en) | Electrocardiogram heart beat classification method based on QRS wave group indefinite cycle superposition | |
Rajput et al. | On arrhythmia detection by deep learning and multidimensional representation | |
Afsar et al. | Detection of ST segment deviation episodes in ECG using KLT with an ensemble neural classifier | |
Srinivas et al. | Adaptive learning based heartbeat classification | |
Zhang et al. | A deep Bayesian neural network for cardiac arrhythmia classification with rejection from ECG recordings | |
Ji et al. | Heartbeat classification based on multifeature combination and stacking-DWKNN algorithm | |
Lamba et al. | A survey of algorithms for feature extraction and feature classification methods |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
EXSB | Decision made by sipo to initiate 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 |
Granted publication date: 20180116 Termination date: 20210522 |
|
CF01 | Termination of patent right due to non-payment of annual fee |