CN116784860A - Electrocardiosignal characteristic extraction system based on morphological heart beat template clustering - Google Patents
Electrocardiosignal characteristic extraction system based on morphological heart beat template clustering Download PDFInfo
- Publication number
- CN116784860A CN116784860A CN202310758487.9A CN202310758487A CN116784860A CN 116784860 A CN116784860 A CN 116784860A CN 202310758487 A CN202310758487 A CN 202310758487A CN 116784860 A CN116784860 A CN 116784860A
- Authority
- CN
- China
- Prior art keywords
- qrs
- heart beat
- template
- waveform
- templates
- 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
- 238000000605 extraction Methods 0.000 title claims abstract description 30
- 230000000877 morphologic effect Effects 0.000 title claims abstract description 15
- 238000000718 qrs complex Methods 0.000 claims abstract description 38
- 238000000034 method Methods 0.000 claims abstract description 19
- 238000012545 processing Methods 0.000 claims abstract description 19
- 238000004364 calculation method Methods 0.000 claims abstract description 14
- 238000010224 classification analysis Methods 0.000 claims abstract description 9
- 238000004458 analytical method Methods 0.000 claims abstract description 8
- 230000002708 enhancing effect Effects 0.000 claims abstract description 5
- 238000005070 sampling Methods 0.000 claims description 14
- 208000008589 Obesity Diseases 0.000 claims description 9
- 235000020824 obesity Nutrition 0.000 claims description 9
- 230000000747 cardiac effect Effects 0.000 claims description 8
- 238000001514 detection method Methods 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 3
- 208000024172 Cardiovascular disease Diseases 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000001228 spectrum Methods 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 239000004744 fabric Substances 0.000 description 2
- 230000036541 health Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 210000003205 muscle Anatomy 0.000 description 1
- 230000003183 myoelectrical effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000000241 respiratory effect Effects 0.000 description 1
- 230000000284 resting effect Effects 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/318—Heart-related electrical modalities, e.g. electrocardiography [ECG]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/318—Heart-related electrical modalities, e.g. electrocardiography [ECG]
- A61B5/346—Analysis of electrocardiograms
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
- A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Heart & Thoracic Surgery (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Cardiology (AREA)
- Evolutionary Computation (AREA)
- Signal Processing (AREA)
- Physiology (AREA)
- Psychiatry (AREA)
- Mathematical Physics (AREA)
- Fuzzy Systems (AREA)
- Theoretical Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The invention discloses an electrocardiosignal characteristic extraction system based on morphological heart beat template clustering, which comprises the following steps: the device comprises a data acquisition module, an extraction module, a data processing module and a classification analysis module; the data acquisition module is used for acquiring electrocardiographic data in a motion state; the extraction module is used for extracting the QRS complex based on the electrocardiograph data to obtain an extracted signal, enhancing the extracted signal and extracting an R peak of the enhanced extracted signal; the data processing module is used for processing the extracted signals based on the R peak to obtain QRS complex heart beats and heart beat characteristic parameters; the classification analysis module is used for classifying the QRS complex heart beats based on the heart beat characteristic parameters to obtain a template library; and carrying out matching analysis on the target heart beat waveform and templates in the template library to obtain an updated template library, and processing electrocardiosignals based on the updated template library. The method has the advantages of high time efficiency of extracting the characteristic value and no need of consuming a large amount of time and calculation resources.
Description
Technical Field
The invention belongs to the technical fields of medical and health, biomedicine and electronic information, and particularly relates to an electrocardiosignal characteristic extraction system based on morphological heart beat template clustering.
Background
Cardiovascular disease is an important disease that severely threatens human health, the prevalence is rising year by year and the diseased population is getting younger and younger, and cardiovascular disease treatment is very complex, requiring long-term monitoring and prevention. An electrocardiogram is one of the common clinical detection modes for diagnosing cardiovascular diseases, and a user needs to lie on a bed in a static state during measurement so as to avoid interference of various noises on electrocardiosignals. However, due to randomness and uncontrollable cardiovascular diseases, the short-time long electrocardiogram of a hospital can only monitor the change condition of heart activity in a short time, and whether the electrocardiosignals are abnormal or not cannot be accurately judged, so that the electrocardiosignals need to be dynamically and stably monitored for a long time. Along with the development of wearable equipment, the monitoring of the dynamic electrocardiogram is mainly completed in the wearable equipment, and the electrocardio electrode can adopt a more comfortable fabric electrode to replace a disposable electrocardio electrode patch, but because the fabric electrode has larger impedance and cannot be firmly fixed on human skin, the electrode-skin impedance change is easily caused by friction with the skin in the movement process, and muscle movement exists during dynamic monitoring, so that the wearable dynamic electrocardiogram is distorted to lose the original information during the monitoring. Based on the above-mentioned problems, a new recognition algorithm for extracting electrocardiosignal information in a motion state is needed.
Disclosure of Invention
The invention aims to solve the defects of the prior art and provides an electrocardiosignal characteristic extraction system based on morphological heart beat template clustering so as to accurately extract the characteristics of electrocardiosignals of a human body in a motion state.
In order to achieve the above object, the present invention provides the following solutions: electrocardiosignal characteristic extraction system based on morphology heart beat template clustering comprises: the device comprises a data acquisition module, an extraction module, a data processing module and a classification analysis module;
the data acquisition module is used for acquiring electrocardiographic data in a motion state;
the extraction module is connected with the data acquisition module and is used for extracting a QRS complex based on the electrocardiograph data to obtain an extraction signal, enhancing the extraction signal and extracting an R peak of the enhanced extraction signal;
the data processing module is connected with the extraction module and is used for processing the extracted signals based on the R peak to obtain QRS wave group heart beats and heart beat characteristic parameters;
the classification analysis module is connected with the data processing module and is used for classifying the QRS complex heart beats based on the heart beat characteristic parameters to obtain a template library; and carrying out matching analysis on the target heart beat waveform and templates in the template library to obtain an updated template library, and processing electrocardiosignals based on the updated template library.
Preferably, the heart beat characteristic parameters include: QRS wave width, QRS wave area, QRS wave height, QRS wave obesity index, and QRS wave peak coarse-pitch index.
Preferably, the QRS wave obesity index calculating method includes:
F QRS ＝A QRS /|H QRS |
wherein F is QRS Representing the QRS wave obesity index, A QRS Represents the QRS wave area, H QRS Representing QRS complex height.
Preferably, the QRS wave peak coarse-ton index calculating method includes:
wherein AG QRS Represents the peak coarse-ton index of the QRS wave, f (n) represents the sampling value of the nth sampling point of the electrocardiosignal, T R Represents the R peak value point position, MS20 represents the number of corresponding sampling points after 20 milliseconds, H QRS Representing QRS complex height.
Preferably, the method for obtaining the template library comprises the following steps: classifying the QRS complex heart beats according to the heart beat characteristic parameters to obtain QRS heart beat templates under three motion states; and carrying out hierarchical clustering on the electrocardiograph data, wherein the QRS heart beat templates respectively correspond to the QRS heart beat templates under three motion states, and the QRS heart beat templates form the template library.
Preferably, the method for obtaining the QRS heart beat template under three motion states comprises the following steps:
defining each QRS complex heart beat as a cluster, and calculating the difference value of all the QRS complex heart beats according to the heart beat characteristic parameters; combining the two clusters with the smallest difference value into a new cluster; calculating dissimilarity between the new cluster and other clusters, updating a similarity matrix based on the dissimilarity, and carrying out the next iteration until the preset iteration times are finished, so as to obtain the QRS heart beat template under three motion states.
Preferably, the method for performing the matching analysis comprises:
constructing an upper limit and a lower limit of a contour waveform by taking a reference waveform in a QRS heart beat template as a center to form a contour window for waveform detection;
aligning a target heart beat waveform with a waveform in the QRS heart beat template in an R peak position, and calculating a difference value of the target heart beat waveform and the waveform in the QRS heart beat template at a time point in the contour window; dividing the difference value at the time point by the sum of the difference values of the P wave, R wave and T wave peaks of the waveforms in the target heart beat waveform and the QRS heart beat template; and obtaining a QRS complex heart beat difference value, wherein the QRS complex heart beat difference value is smaller than 0.7, and the matching is successful.
Preferably, the method for obtaining the updated template library comprises the following steps:
when the waveform in the QRS heart beat template has a waveform matched with the target heart beat waveform, adding the target heart beat waveform into the QRS heart beat template, and updating the QRS heart beat template to obtain the updated template library;
checking whether the waveforms in the QRS heart beat templates are 8 or not when the waveforms in the QRS heart beat templates are not matched with the waveforms in the target heart beat templates, and when the waveforms in the QRS heart beat templates are not 8, taking the target heart beat waveforms as prototypes, establishing a new template in the QRS heart beat templates, and setting the number of the target heart beat waveform templates as the number of the new template; when the number of the QRS heart beat templates reaches 8, deleting one template in the QRS heart beat templates, taking the target heart beat waveform as a prototype, establishing a new template in the QRS heart beat templates, and setting the number of the target heart beat waveform template as the number of the new template.
Compared with the prior art, the invention has the beneficial effects that:
the electrocardiosignal characteristic extraction system based on morphological heart beat template clustering has the advantages of high time efficiency for extracting characteristic values and no need of consuming a large amount of time and calculation resources. (1) The precision is high, fine structures and related features in the electrocardiosignal can be accurately captured, and the precision of identification and analysis of the electrocardiosignal is improved; (2) robust and interpreted: compared with a neural network algorithm, the morphological heart beat template clustering algorithm has stronger robustness and interpretation, and can better describe the characteristics, trend and abnormal condition of electrocardiosignals; (3) high interpretability: the diagnosis result can be interpreted through the specific form and parameter characteristics of the heartbeat template, and the diagnosis result is more visual and easy to understand and realize.
Drawings
In order to more clearly illustrate the technical solutions of the present invention, the drawings that are needed in the embodiments are briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of a method for matching analysis by a classification analysis module according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a template library portion in a stationary state according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a portion of a template library sample in a routine state of the present invention;
FIG. 4 is a schematic diagram of a portion of a template library in a running state according to an embodiment of the present invention;
FIG. 5 is a graph of an electrocardiographic signal spectrum in a resting state according to an embodiment of the present invention;
FIG. 6 is a graph of an electrocardiographic signal after noise signal removal in a stationary state according to an embodiment of the present invention;
FIG. 7 is a graph of an electrocardiographic signal spectrum in a routine of the present invention;
FIG. 8 is a graph of an electrocardiographic signal after noise signal removal in a routine state of the present invention;
FIG. 9 is a graph of the electrocardiographic signal spectrum in the running state according to the embodiment of the present invention;
fig. 10 is a graph of an electrocardiographic signal after removing noise signals in a running state according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In order that the above-recited objects, features and advantages of the present invention will become more readily apparent, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description.
Example 1
The embodiment provides an electrocardiosignal characteristic extraction system based on morphological heart beat template clustering, which comprises the following steps: the device comprises a data acquisition module, an extraction module, a data processing module and a classification analysis module;
the data acquisition module is used for acquiring electrocardiographic data in a motion state.
In this embodiment, the electrocardiographic data includes: electrocardiogram data f collected in three motion states of static, running (10 km/h) and walking (4 km/h) 1(x) 、f 2(x) 、f 3(x) 。
The extraction module is connected with the data acquisition module and is used for extracting the QRS complex based on the electrocardiograph data to obtain an extraction signal, enhancing the extraction signal and extracting an R peak of the enhanced extraction signal.
In this embodiment, the Patrick s.hamilton algorithm is used to extract QRS complexes of the electrocardiographic data under the three motion states, and the extracted signals after extraction are f respectively 1QRS(x) 、f 2QRS(x) 、f 3QRS(x) . And then enhancing by adopting a window sliding integral function, and extracting R peaks of the enhanced extracted signals by adopting a double-slope variable threshold method.
The data processing module is connected with the extraction module and is used for processing the extracted signals based on the R peak to obtain QRS complex heart beats and heart beat characteristic parameters.
The processing method comprises the following steps:
the extracted signal firstly passes through a 16Hz low-pass filter and then passes through an 8Hz high-pass filter to filter out respiratory interference, power frequency interference and myoelectric noise; the baseline shift is then removed by a median filter and the electrocardiographic signal baseline is pulled to zero potential. The purpose of using strong filtering (8-16 Hz) in the preprocessing is to maximize the QRS wave energy, while pulling the baseline to zero potential is to facilitate calculation of the characteristic parameters such as the subsequent burst area.
Specifically, the start point T of each QRS complex heart beat waveform is firstly determined according to the extracted R peak position 0 And end point T s Calculating characteristic parameters of the heart beat; the heart beat characteristic parameters of the embodiment comprise: QRS wave width (W QRS ) QRS wave area (a) QRS ) QRS wave height (H) QRS ) QRS wave obesity index (F QRS ) And QRS peak coarse-to-peak index (AG QRS )。
Wherein the QRS wave width (W QRS ) The calculation method comprises the following steps:
W QRS ＝T s -T 0
in which W is QRS Representing the QRS wave width, T s Represents the end point, T, of the QRS complex heart beat waveform 0 Representing the start of the QRS complex heart beat waveform.
QRS wave area (a QRS ) The calculation method comprises the following steps: origin T of QRS complex heart beat waveform 0 Starting to integrate absolute value of the preprocessed electrocardio data sampling value to the end point T of the QRS complex heart beat waveform s And (5) ending. The calculation formula is as follows:
wherein A is QRS Representing the area of the QRS wave, f (n) representing the sampled value of the nth sampling point of the electrocardiosignal, T Q Representing the start of the QRS wave.
QRS wave height (H QRS ) The calculation method comprises the following steps: r peak value point position (T) R ) Positive values indicate the positive R-peak direction and negative values indicate the negative R-peak direction of the sample values of the pre-processed electrocardiographic data. The calculation formula is as follows:
H QRS ＝f(T R )
wherein H is QRS Representing the height of the QRS wave, T R Representing the R-peak value point location.
QRS wave obesity index (F QRS ) The calculation method comprises the following steps: the ratio of QRS complex heart beat waveform area to the absolute value of height. The calculation formula is as follows:
F QRS ＝A QRS /|H QRS |
wherein F is QRS Representing the QRS wave obesity index, A QRS Represents the QRS wave area, H QRS Representing QRS complex height.
QRS wave peak coarse-ton index (AG QRS ) The calculation method of (2) is as follows:
wherein AG QRS Representing QRS wave peak coarse-pitchAn index f (n) represents the sampling value of the nth sampling point of the electrocardiosignal, T R Represents the R peak value point position, MS20 represents the number of corresponding sampling points after 20 milliseconds, H QRS Representing QRS complex height.
The classification analysis module is connected with the data processing module and is used for classifying the QRS complex heart beats based on the heart beat characteristic parameters to obtain a template library; and carrying out matching analysis on the target heart beat waveform and templates in the template library to obtain an updated template library, and processing electrocardiosignals based on the updated template library.
Specifically, the method for obtaining the template library comprises the following steps: classifying the QRS complex cardiac beats according to the five cardiac beat characteristic parameters, and classifying all cardiac beats by adopting a hierarchical clustering algorithm to obtain QRS cardiac beat templates under three motion states; specifically, the QRS complex heart beat is defined as a cluster, and the difference values of all the QRS complex heart beats are calculated according to the heart beat characteristic parameters;
QRS complex cardiac beat difference value S DIFF The calculation method comprises the following steps:
wherein T (n) is the sampling value of the nth sampling point of the template heart beat waveform data, X (n) is the sampling value of the nth sampling point of the target heart beat waveform data, R peak For R peak position, K D Is constant, T max (T min ) For the template heart beat waveform data in [ R ] peak -K D ,R peak +K D ]Maximum (small) value in range, X max (X min ) For the target heart beat waveform data in [ R ] peak -K D ,R peak +K D ]Maximum (small) value within the range.
Combining the two clusters with the smallest difference value into a new cluster; calculating dissimilarity between the new cluster and other clusters, updating a similarity matrix based on the dissimilarity, and carrying out the next iteration until the preset iteration times are finished. And obtaining QRS heart beat templates under three motion states. Hierarchical clustering is carried out on the electrocardiographic data, and the electrocardiographic data f 1(x) 、f 2(x) 、f 3(x) Obtaining P after hierarchical clustering 11(x) ，P 12(x) ，…，P 21(x) ，P 22(x) ，…，P 31(x) ，P 32(x) … respectively correspond to the QRS heart beat templates S under the three motion states 1 、S 2 、S 3 QRS heart beat templates form a template library. As shown in fig. 2,3 and 4.
The method for matching and analyzing the electrocardio data and the templates in the template library comprises the following steps:
in the embodiment, an outline limiting and accumulated difference method is adopted to match the electrocardiographic data with templates in a template library; the contour limit refers to the waveform in the QRS heart beat template as the reference waveform of the contour limit, and the upper limit deltaf of the contour waveform is constructed by taking the reference waveform as the center 1 And a lower limit Δf 2 Form a contour window of waveform detection, i.e. the contour window is x=f r(x) +Δf 1 -Δf 2 Wherein, r takes the values of 1,2 and 3. The cumulative difference method is to pair the target heart beat waveform and the waveform in the QRS heart beat template at the R peak position, and then calculate the difference value of the target waveform and the waveform in the QRS heart beat template at each time point within the range of the contour window (generally the width of the QRS wave is 60-100 ms); dividing the difference value at the time point by the sum of the difference values of the P wave, the R wave and the T wave peaks of the target heart beat waveform and the waveform in the QRS heart beat template to obtain a QRS wave group heart beat difference value S DIFF 。
In this embodiment, the judging conditions for successful matching of the target heart beat waveform and the waveform in the QRS heart beat template include:
(1) The R peak is in the same direction;
(2) The body types are the same;
(3)|H QRS (X)-H QRS (T)|<|H QRS (X)+H QRS (T) |/8, wherein H QRS (X) is the QRS wave height, H of the target heart beat waveform QRS (T) is the QRS complex height of the waveform in the QRS heart beat template;
(4)|W QRS (X)-W QRS (T)|≤20ms，W QRS (X) is the QRS wave width, W of the target heart beat waveform QRS (T) is a wave in a QRS heart beat templateQRS wave width of shape;
(5)S DIFF <0.7。
specifically, as shown in fig. 1, the method for obtaining the updated template library includes:
when the waveform in the QRS heart beat template has a waveform matched with the target heart beat waveform, the target heart beat waveform is represented by P mi(x) (m is the type of the QRS heart beat template, i is the number of signals in the QRS heart beat template), and adding the template into the QRS heart beat template, and updating the QRS heart beat template to obtain an updated template library.
Checking whether the waveforms in the QRS heart beat templates reach 8 when the waveforms in the QRS heart beat templates do not have the waveforms matched with the target heart beat waveforms, when the waveforms do not reach 8, taking the target heart beat waveforms as prototypes, building a new template in the QRS heart beat templates, and setting the number of the target heart beat waveform templates as the number S of the new template n (n is more than or equal to 4 and less than or equal to 8); when the number of the QRS heart beat templates reaches 8, deleting one of the templates, taking the target heart beat waveform as a prototype, establishing a new template in the QRS heart beat template, and setting the number of the target heart beat waveform template as the number of the new template. And clustering the electrocardiosignals by using the updated template library. As can be seen in FIGS. 5-10, the invention can effectively remove the motion noise of the electrocardiosignal.
The above embodiments are merely illustrative of the preferred embodiments of the present invention, and the scope of the present invention is not limited thereto, but various modifications and improvements made by those skilled in the art to which the present invention pertains are made without departing from the spirit of the present invention, and all modifications and improvements fall within the scope of the present invention as defined in the appended claims.
Claims (8)
1. Electrocardiosignal characteristic extraction system based on morphology heart beat template clustering, which is characterized by comprising: the device comprises a data acquisition module, an extraction module, a data processing module and a classification analysis module;
the data acquisition module is used for acquiring electrocardiographic data in a motion state;
the extraction module is connected with the data acquisition module and is used for extracting a QRS complex based on the electrocardiograph data to obtain an extraction signal, enhancing the extraction signal and extracting an R peak of the enhanced extraction signal;
the data processing module is connected with the extraction module and is used for processing the extracted signals based on the R peak to obtain QRS wave group heart beats and heart beat characteristic parameters;
the classification analysis module is connected with the data processing module and is used for classifying the QRS complex heart beats based on the heart beat characteristic parameters to obtain a template library; and carrying out matching analysis on the target heart beat waveform and templates in the template library to obtain an updated template library, and processing electrocardiosignals based on the updated template library.
2. The system for extracting cardiac signal features based on morphological cardiac beat template clustering according to claim 1, wherein the cardiac beat feature parameters comprise: QRS wave width, QRS wave area, QRS wave height, QRS wave obesity index, and QRS wave peak coarse-pitch index.
3. The system for extracting electrocardiographic features based on morphological heart beat template clustering according to claim 2, wherein the QRS wave obesity index calculation method comprises:
F QRS ＝A QRS /|H QRS |
wherein F is QRS Representing the QRS wave obesity index, A QRS Represents the QRS wave area, H QRS Representing QRS complex height.
4. The system for extracting electrocardiographic features based on morphological heart beat template clustering according to claim 2, wherein the QRS wave peak coarse-to-ton index calculation method comprises:
wherein AG QRS Representing QRS wave spike coarseThe ton index, f (n), represents the sampled value of the nth sampling point of the electrocardiosignal, T R Represents the R peak value point position, MS20 represents the number of corresponding sampling points after 20 milliseconds, H QRS Representing QRS complex height.
5. The system for extracting electrocardiographic features based on morphological heart beat template clustering according to claim 1, wherein the method for obtaining the template library comprises: classifying the QRS complex heart beats according to the heart beat characteristic parameters to obtain QRS heart beat templates under three motion states; and carrying out hierarchical clustering on the electrocardiograph data, wherein the QRS heart beat templates respectively correspond to the QRS heart beat templates under three motion states, and the QRS heart beat templates form the template library.
6. The system for extracting electrocardiographic signal features based on morphological heart beat template clustering according to claim 5, wherein the method for obtaining QRS heart beat templates under three motion states comprises:
defining each QRS complex heart beat as a cluster, and calculating the difference value of all the QRS complex heart beats according to the heart beat characteristic parameters; combining the two clusters with the smallest difference value into a new cluster; calculating dissimilarity between the new cluster and other clusters, updating a similarity matrix based on the dissimilarity, and carrying out the next iteration until the preset iteration times are finished, so as to obtain the QRS heart beat template under three motion states.
7. The system for extracting electrocardiographic features based on morphological heart beat template clustering according to claim 5, wherein the method for performing matching analysis comprises:
constructing an upper limit and a lower limit of a contour waveform by taking a reference waveform in a QRS heart beat template as a center to form a contour window for waveform detection;
aligning a target heart beat waveform with a waveform in the QRS heart beat template in an R peak position, and calculating a difference value of the target heart beat waveform and the waveform in the QRS heart beat template at a time point in the contour window; dividing the difference value at the time point by the sum of the difference values of the P wave, R wave and T wave peaks of the waveforms in the target heart beat waveform and the QRS heart beat template; and obtaining a QRS complex heart beat difference value, wherein the QRS complex heart beat difference value is smaller than 0.7, and the matching is successful.
8. The system for extracting electrocardiographic features based on morphological heart beat template clustering according to claim 7, wherein the method for obtaining the updated template library comprises:
when the waveform in the QRS heart beat template has a waveform matched with the target heart beat waveform, adding the target heart beat waveform into the QRS heart beat template, and updating the QRS heart beat template to obtain the updated template library;
checking whether the waveforms in the QRS heart beat templates are 8 or not when the waveforms in the QRS heart beat templates are not matched with the waveforms in the target heart beat templates, and when the waveforms in the QRS heart beat templates are not 8, taking the target heart beat waveforms as prototypes, establishing a new template in the QRS heart beat templates, and setting the number of the target heart beat waveform templates as the number of the new template; when the number of the QRS heart beat templates reaches 8, deleting one template in the QRS heart beat templates, taking the target heart beat waveform as a prototype, establishing a new template in the QRS heart beat templates, and setting the number of the target heart beat waveform template as the number of the new template.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310758487.9A CN116784860B (en) | 2023-06-26 | 2023-06-26 | Electrocardiosignal characteristic extraction system based on morphological heart beat template clustering |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310758487.9A CN116784860B (en) | 2023-06-26 | 2023-06-26 | Electrocardiosignal characteristic extraction system based on morphological heart beat template clustering |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116784860A true CN116784860A (en) | 2023-09-22 |
| CN116784860B CN116784860B (en) | 2024-05-28 |
Family
ID=88035904
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310758487.9A Active CN116784860B (en) | 2023-06-26 | 2023-06-26 | Electrocardiosignal characteristic extraction system based on morphological heart beat template clustering |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116784860B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117357130A (en) * | 2023-12-07 | 2024-01-09 | 深圳泰康医疗设备有限公司 | Electrocardiogram digital curve segmentation method based on artificial intelligence |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140128758A1 (en) * | 2012-11-08 | 2014-05-08 | Conner Daniel Cross Galloway | Electrocardiogram signal detection |
| CN111281373A (en) * | 2020-03-06 | 2020-06-16 | 何乐 | Method and device for quantitatively evaluating cardiac function based on electrocardiogram U wave and T wave |
| CN115470832A (en) * | 2022-11-14 | 2022-12-13 | 南京邮电大学 | Electrocardiosignal data processing method based on block chain |
| US20230277111A1 (en) * | 2020-07-16 | 2023-09-07 | Sridhar Krishnan | System and method for saliency detection in long-term ecg monitoring |
-
2023
- 2023-06-26 CN CN202310758487.9A patent/CN116784860B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140128758A1 (en) * | 2012-11-08 | 2014-05-08 | Conner Daniel Cross Galloway | Electrocardiogram signal detection |
| CN111281373A (en) * | 2020-03-06 | 2020-06-16 | 何乐 | Method and device for quantitatively evaluating cardiac function based on electrocardiogram U wave and T wave |
| US20230277111A1 (en) * | 2020-07-16 | 2023-09-07 | Sridhar Krishnan | System and method for saliency detection in long-term ecg monitoring |
| CN115470832A (en) * | 2022-11-14 | 2022-12-13 | 南京邮电大学 | Electrocardiosignal data processing method based on block chain |
Non-Patent Citations (1)
| Title |
|---|
| 曹雪滨等: "15例扩张型心肌病心室晚电位分析", 中国保健医学研究会心脏学学会全国第一届心脏学学术会议论文汇编, 1 October 1995 (1995-10-01) * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117357130A (en) * | 2023-12-07 | 2024-01-09 | 深圳泰康医疗设备有限公司 | Electrocardiogram digital curve segmentation method based on artificial intelligence |
| CN117357130B (en) * | 2023-12-07 | 2024-02-13 | 深圳泰康医疗设备有限公司 | Electrocardiogram digital curve segmentation method based on artificial intelligence |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116784860B (en) | 2024-05-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Hagiwara et al. | Computer-aided diagnosis of atrial fibrillation based on ECG Signals: A review | |
| AL-Ziarjawey et al. | Heart rate monitoring and PQRST detection based on graphical user interface with Matlab | |
| Karpagachelvi et al. | ECG feature extraction techniques-a survey approach | |
| Gupta et al. | A comparison of ECG signal pre-processing using FrFT, FrWT and IPCA for improved analysis | |
| Oweis et al. | QRS detection and heart rate variability analysis: A survey | |
| JP7429371B2 (en) | Method and system for quantifying and removing asynchronous noise in biophysical signals | |
| Li et al. | Dual-input neural network integrating feature extraction and deep learning for coronary artery disease detection using electrocardiogram and phonocardiogram | |
| CN114052744B (en) | Electrocardiosignal classification method based on impulse neural network | |
| CN112971795B (en) | Electrocardiosignal quality evaluation method | |
| CN116784860B (en) | Electrocardiosignal characteristic extraction system based on morphological heart beat template clustering | |
| Jaswal et al. | QRS detection using wavelet transform | |
| Talatov et al. | Algorithmic and software analysis and processing of ECG signals | |
| CN111832537A (en) | Abnormal electrocardiosignal identification method and abnormal electrocardiosignal identification device | |
| Hugeng et al. | Development of the ‘Healthcor’system as a cardiac disorders symptoms detector using an expert system based on arduino uno | |
| CN113609975A (en) | Modeling method for tremor detection, hand tremor detection device and method | |
| Zaidi et al. | Feature extraction and characterization of cardiovascular arrhythmia and normal sinus rhythm from ECG signals using LabVIEW | |
| Dogan et al. | A comprehensive review of computer-based Techniques for R-peaks/QRS complex detection in ECG signal | |
| CN115486854A (en) | Single lead electrocardiogram ventricular premature beat identification method aiming at dry electrode acquisition | |
| CN115024716A (en) | Ballistocardiogram signal reconstruction method based on heart rate label generation | |
| Dharma et al. | Hypertension Identification Using Naive Bayes Classification Method and Pan Tompkins Feature Extraction | |
| Wirananta et al. | Prediction of Atrial Fibrillation using XGBoost and Smartphone PPG Data with ECG Based Model | |
| Sanamdikar et al. | Analysis of several characteristics of ECG signal for cardiac arrhythmia detection | |
| Aravind et al. | ECG Classification and Arrhythmia Detection Using Wavelet Transform and Convolutional Neural Network | |
| Kammath et al. | Detection of Bundle Branch Blocks using Machine Learning Techniques | |
| Nankani et al. | R-peak detection from ECG signals using fractal based mathematical morphological operators |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |