CN114027853A - QRS complex detection method, device, medium and equipment based on feature template matching - Google Patents
QRS complex detection method, device, medium and equipment based on feature template matching Download PDFInfo
- Publication number
- CN114027853A CN114027853A CN202111542679.3A CN202111542679A CN114027853A CN 114027853 A CN114027853 A CN 114027853A CN 202111542679 A CN202111542679 A CN 202111542679A CN 114027853 A CN114027853 A CN 114027853A
- Authority
- CN
- China
- Prior art keywords
- detector
- feature
- template
- qrs complex
- qrs
- 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
Images
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]
- A61B5/346—Analysis of electrocardiograms
- A61B5/349—Detecting specific parameters of the electrocardiograph cycle
- A61B5/366—Detecting abnormal QRS complex, e.g. widening
-
- 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
-
- 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
- A61B5/7253—Details of waveform analysis characterised by using transforms
-
- 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
- A61B5/7253—Details of waveform analysis characterised by using transforms
- A61B5/7257—Details of waveform analysis characterised by using transforms using Fourier transforms
Abstract
The invention discloses a QRS complex detection method, a device, a medium and equipment based on feature template matching, and belongs to the technical field of electrocardio. Firstly, preprocessing a electrocardiosignal, and then performing characteristic conversion by using the preprocessed signal; then, a reliable classical detection algorithm is selected to carry out strict threshold screening on the QRS wave complex, so that the false alarm rate of the QRS wave complex is extremely low; then, making a multi-class characteristic template by using the screened QRS wave group; then, a QRS complex to be matched is screened by using a classical detection algorithm for reducing a threshold value, and an extremely low missing report rate is kept; and finally, performing similarity matching calculation on the candidate QRS complexes by using the multi-class feature template, performing threshold division, and resampling the QRS complexes in the result to the original frequency. The invention adopts a method of combining a plurality of QRS complex templates and a plurality of waveform characteristics, and effectively improves the detection speed while keeping higher detection accuracy compared with the prior art.
Description
Technical Field
The invention belongs to the technical field of electrocardio, and particularly relates to a QRS complex detection method, a QRS complex detection device, a QRS complex detection medium and QRS complex detection equipment based on feature template matching.
Background
In the field of medical health, cardiovascular diseases as a common chronic and easily-emergent disease always troubles a large number of patients and medical workers, and for most cardiovascular chronic diseases, a medical technology capable of completely curing the cardiovascular chronic diseases does not exist at present. Because cardiovascular diseases are mostly chronic diseases and have higher outbreak probability, the ideal method is to carry out all-weather medical monitoring on patients, but common patients cannot bear the high cost of receiving long-term monitoring in hospitals, the diseases belong to the chronic diseases, the physical signs of the patients are completely the same as those of healthy people when the patients do not suffer from the diseases, and if the long-term hospitalization monitoring is adopted, the waste of medical resources is also really caused. Therefore, a simple and easy-to-operate method for real-time dynamic monitoring of the heart condition of a patient is a necessary means. The electrocardiographic examination technology is a non-invasive examination means widely applied to heart disease diagnosis at present and is a main basis for evaluating and judging the health condition of the heart.
In the context of industrial automation, the basis of automatic analysis and diagnosis systems is the detection of characteristic parameters of the cardiac electrical signal and the waveform recognition. The basis of automatic analysis of the electrocardiosignals is accurate identification of QRS waves in the electrocardiosignals, and then R waves in a QRS wave group are identified, so that correct heart rate calculation, heart rate variability analysis, ST-segment parameter detection and correct normal and abnormal heart rate distinguishing can be guaranteed.
At present, the main methods include a difference threshold method, a template matching method, a neural network method, a wavelet transform method, and the like. The principle of the differential threshold method is relatively simple, the calculation speed is high, normal electrocardiosignals can be effectively detected, and the detection effect on signals containing noise interference is not ideal; the template matching method is a detection algorithm based on statistical identification, is relatively stable, but has high repeated execution rate and relatively long time consumption; the neural network method is to input the electrocardiogram waveforms with different characteristics into a back propagation neural network for identification, has obvious detection effect, but needs a large amount of time for processing data; the wavelet transform method is characterized in that the R wave of the electrocardiosignal has strong singular characteristics, the wavelet transform is utilized to perform correlation analysis on the electrocardiosignal, and then the shape of the input signal is compared with a wavelet template function.
In order to improve the accuracy of QRS wave detection, a template matching detection method based on a difference threshold method is proposed in the prior art. However, this solution still has some drawbacks: firstly, the selection of the template is too single, and normal and abnormal QRS waves cannot be considered; secondly, the selected template has single characteristic and cannot effectively represent the characteristic of the signal; thirdly, after the template is selected, sliding window comparison needs to be carried out on all signals, and the calculated amount is high.
Disclosure of Invention
The technical problem is as follows: aiming at the problems in the prior art, the invention provides a QRS complex detection method, a device, a medium and equipment based on feature template matching.
The technical scheme is as follows: in a first aspect, the present invention provides a QRS complex detection method based on feature template matching, including:
receiving an electrocardiosignal;
preprocessing the electrocardiosignals, comprising: resampling the electrocardiosignals to a fixed frequency, and filtering;
performing characteristic conversion on the preprocessed signals;
acquiring a strict position of a template QRS complex by utilizing at least one first detector and adjusting a threshold value of the first detector; wherein the threshold of the first detector is greater than or equal to a first set value;
cutting heartbeat feature fragments of the preprocessed data and the data after feature conversion by taking the QRS wave group position of the template as a center, and dividing different feature fragments into a plurality of groups by using a similarity function to form a plurality of groups of feature templates;
screening out candidate QRS complex positions through at least one second detector; wherein the threshold of the second detector is less than or equal to a second set value;
calculating the similarity between the candidate QRS wave group and each group of characteristic templates, wherein the maximum value of the similarity is selected in the same group as a result, and different parameter strategies are selected among different groups;
and selecting a final result according to a set threshold value, and resampling to an original frequency.
Further, the performing feature conversion on the preprocessed signal includes:
and performing feature transformation by adopting any one or any combination of more of derivatives, differences, Fourier transform, wavelet transform, Hilbert transform and information entropy.
Further, the first detector is any one or any combination of open source detection algorithms of pantompkins1985, hamilton2002, martinez2003, christov2004, gamboa2008, elgendi2010, engzeemod2012, kallidas 2017, rodrigues2021, promac, wqrs, gqrs and xqrs.
Further, acquiring, with at least one detector, a strict template QRS complex position by adjusting a threshold of a first detector comprises:
respectively giving weight to each first detector, and calculating a first detection probability of a certain point according to whether the first detector detects the certain point and the weight corresponding to the first detector;
and comparing the first detection probability with a threshold value of a first detector, and determining the position of the QRS complex of the template according to the comparison result.
Further, the dividing the different feature segments into a plurality of groups using the similarity function, and the forming a plurality of groups of feature templates includes:
calculating the average value of the integral characteristic segment on each dimension as a standard template;
respectively calculating the similarity of each characteristic data segment and the standard template;
removing characteristic data segments with large difference with the identification degree of the standard template, sequencing the rest similarity, and dividing the characteristic data segments into a plurality of groups by taking a place with a high similarity change rate as a boundary of the groups;
and respectively calculating the average value of each group of characteristic data fragments on each dimension to obtain a characteristic data template. Further, the second detector is any one or any combination of open source detection algorithms of pantompkins1985, hamilton2002, martinez2003, christov2004, gamboa2008, elgendi2010, engzeemod2012, kallidas 2017, rodrigues2021, promac, wqrs, gqrs and xqrs.
Further, the method for screening out the candidate QRS complex position by the second detector comprises:
respectively giving a weight to each second detector, and calculating a second detection probability of a certain point according to whether the second detector detects the point and the weight corresponding to the second detector;
and comparing the second detection probability with a threshold value of a basic detector, and determining the position of the candidate QRS complex according to the comparison result.
In a second aspect, the present invention provides a QRS complex detection apparatus based on feature template matching, which detects a QRS complex according to any one of the QRS complex detection methods based on feature template matching proposed by the present invention, and is characterized by comprising:
a signal receiving unit configured to receive an electrocardiographic signal;
a preprocessing unit configured to preprocess the cardiac electrical signal, comprising: resampling the electrocardiosignals to a fixed frequency, and filtering;
a feature conversion unit configured to perform feature conversion on the preprocessed signal;
the first screening unit is used for acquiring a strict template QRS complex position by utilizing at least one first detector and adjusting a threshold value of the first detector; wherein the threshold of the base detector is greater than or equal to a first set value;
the template generating unit is configured to cut heartbeat feature segments of the preprocessed data and the feature-converted data by taking the QRS complex position of the template as a center, and divide different feature segments into a plurality of groups by using a similarity function to form a plurality of groups of feature templates;
a second screening unit configured to screen out candidate QRS complex positions by at least one second detector; wherein the threshold of the second detector is less than or equal to a second set value;
a calculating unit configured to calculate similarity between the candidate QRS complex and each set of feature templates; wherein, the maximum value of the similarity is selected in the same group, and different parameter strategies are selected among different groups;
and the third screening unit is configured to select a final result according to a set threshold value and resample the final result to the original frequency.
In a third aspect, the present invention provides a storage medium, in which computer instructions are stored, and when the computer is executed, the method for QRS complex detection based on feature template matching according to any one of the present invention can be executed.
In a fourth aspect, the present invention provides an electronic device comprising:
a storage medium proposed by the present invention; and the number of the first and second groups,
a processor capable of executing computer instructions stored in the storage medium.
Compared with the prior art, the method has the advantages that firstly, electrocardiosignals are preprocessed, and then the preprocessed signals are used for feature conversion; then, a reliable classical detection algorithm is selected to carry out strict threshold screening on the QRS wave complex, so that the false alarm rate of the QRS wave complex is extremely low; then, making a multi-class characteristic template by using the screened QRS wave group; then, a QRS complex to be matched is screened by using a classical detection algorithm for reducing a threshold value, and an extremely low missing report rate is kept; and finally, performing similarity matching calculation on the candidate QRS complexes by using the multi-class feature template, performing threshold division, and resampling the QRS complexes in the result to the original frequency. The invention adopts a method of combining a plurality of QRS complex templates and a plurality of waveform characteristics, and effectively improves the detection speed while keeping higher detection accuracy compared with the prior art.
Drawings
Fig. 1 is a flowchart of a QRS complex detection method based on feature template matching in an embodiment of the present invention;
fig. 2 is a block diagram of a QRS complex detection apparatus based on feature template matching in an embodiment of the present invention;
fig. 3 is a block diagram of an electronic device in an embodiment of the invention.
Detailed Description
The invention is further described with reference to the following examples and the accompanying drawings. It is noted that the terms "first," "second," and the like are used for convenience of description only and are not to be construed as limiting in number or the like.
Fig. 1 shows a flowchart of a QRS complex detection method based on feature template matching in an embodiment of the present invention, and in combination with fig. 1, in an embodiment of the present invention, the method includes:
step S100: receiving the electrocardiosignals.
Step S200: and preprocessing the electrocardiosignals. In an embodiment of the invention, the signal is first resampled to a fixed frequency, for example 500Hz in one example, and sample normalization is performed; a 50Hz notch filter is then used to filter out the power frequency interference. It should be noted that the resampling method is a method existing in the prior art, and this embodiment will not be described in more detail.
Step S300: and performing characteristic conversion on the preprocessed signals. In the embodiment of the invention, any one or any combination of more of derivative, difference, Fourier transform, wavelet transform, Hilbert transform and information entropy is adopted for feature transformation. For example, in one example, the first derivative is used for feature transformation. The first derivative is adopted for feature conversion, and the calculation formula is as follows:
y(n)＝k*[x(n)+2*x(n-1)-2*x(n-3)-x(n-4)]···(1)
in formula (1), x represents a preprocessed signal, and n is 1,2, 3. M represents a signal length; y represents the first derivative of x; when x (i) is absent, 0 is used instead; the formula for k is as follows:
k＝fs|8···(2)
wherein fs represents the sampling rate after preprocessing; k represents the result of fs dividing by 8.
Step S400: and acquiring the position of the strict template QRS complex by utilizing at least one first detector and adjusting the threshold value of the first detector. In this step, the input data for detecting QRS waves may be preprocessed data or data subjected to feature conversion.
In the embodiment of the present invention, currently known open source detection algorithms are selected as the first detector, and these open source detection algorithms are selected to be reliable, for example, any one or any combination of pantomarks 1985, hamilton2002, martinez2003, christov2004, gamboa2008, elgendi2010, engzeemod2012, kallidas 2017, rodrigues2021, promac, wqrs, gqrs, xqrs open source detection algorithms, and a higher threshold is used to obtain the position of the template QRS waveform.
Specifically, each first detector may be given a weight, and a first detection probability of a certain point is calculated according to whether the first detector detects the certain point and the weight corresponding to the first detector; and comparing the first detection probability with a threshold value of the first detector, and determining the position of the QRS complex of the template according to the comparison result.
More specifically, in one example of the present invention, for the selection of the detection results of the plurality of first detectors, the final position of the wave group may be determined by using a method of voting by giving different weights; as in this example, using pantopkins 1985, hamilton2002, and rodrigues2021 as the first detector, the calculation results of pantopkins 1985, hamilton2002, and rodrigues2021 are respectively given weights of 0.4, 0.3, and when the weight of the same point exceeds 0.5, that is, the first set value, the point is regarded as a QRS wave.
The first detection probability expression is:
where r denotes the number of detection algorithms (first detectors) used; i indicates whether the detection algorithm (first detector) detected the point, detected as 1, not detected as 0; w represents the weight corresponding to the detection algorithm (first detector) and has:
screening the expression according to the first detection probability as follows:
wherein, R represents the final screening result, 1 represents QRS wave, and 0 represents not QRS wave; th denotes a threshold value of the first detector.
Step S500: and performing heartbeat feature fragment cutting on the preprocessed data and the data after feature conversion by taking the position of the QRS wave group of the template as a center, and dividing different feature fragments into a plurality of groups by using a similarity function to form a plurality of groups of feature templates. After the screening of step S400, the false positive detection is substantially absent, i.e., the generation influence of the false positive portion on the final template is almost negligible. At this time, the heart beat feature segmentation is performed on the preprocessed data and the feature-converted data with the position of the template QRS complex as the center, for example, in one example, the segmentation is performed at the first 0.25s and the last 0.3s of the position of the template QRS complex.
Considering abnormal morphology of QRS waves, such as ventricular premature beat, the different feature segment templates are divided into 2 groups using similarity functions, respectively.
In the embodiment of the invention, cosine similarity, Pearson correlation coefficient and the like can be adopted to represent similarity, and similarity functions can be rewritten by adding characteristics such as assignment and the like. Wherein the cosine similarity function is expressed as follows:
in formula (6), Similarity represents Similarity; x, Y respectively represent feature data pieces.
The pearson correlation coefficient function is expressed as follows:
in the formula (7), ρ represents the similarity; x, Y respectively represent feature data pieces.
In the embodiment of the present invention, in order to classify different feature segment templates using the similarity function, specifically, the following steps S510 to S540 may be performed:
step S510: calculating the average value of the overall feature segment in each dimension as a standard template, wherein the calculation formula is as follows:
wherein, TiRepresenting the ith segment in the feature segment group; t represents a standard template。
Step S520: and respectively calculating the similarity of each characteristic data segment and the standard template, wherein the cosine similarity and the Pearson correlation coefficient can be adopted.
Step S530: and removing the characteristic data segments with larger difference with the identification degree of the standard template, sequencing the rest similarities, and dividing the characteristic data segments into a plurality of groups by taking the place with the higher similarity change rate as a grouping boundary. In this step, the purpose of rejecting the feature data segment with a large difference in identity with the standard template is to remove interference of partial noise.
Step S540: and respectively calculating the average value of each group of characteristic data fragments on each dimension to obtain a characteristic data template. Specifically, the calculation may be performed according to formula (8) in step S510.
Step S600: and screening out the candidate QRS complex positions through at least one second detector. The main purpose of this step is to screen candidate QRS complex positions in advance by known efficient detectors to reduce the complexity of the overall method.
In the embodiment of the present invention, this step mainly adopts a lower threshold open source algorithm, such as any one or any combination of pantompkins1985, hamilton2002, martinez2003, christov2004, gamboa2008, elgendi2010, engzeemod2012, kallidas 2017, rodrigues2021, promac, wqrs, gqrs and xqrs open source detection algorithms. In one example of the present invention, the second detector is employed using gamboa2008 and kallidas 2017.
Specifically, the implementation of this step is similar to step S400, each second detector is given a weight, and a second detection probability of the point is calculated according to whether the second detector detects the point and the weight corresponding to the second detector; and comparing the second detection probability with a threshold value of a second detector, and determining the position of the candidate QRS complex according to the comparison result. The specific calculation formula can refer to formulas (3) - (5), and the calculation can be completed by replacing the meanings of the corresponding letters. In an embodiment of the invention, the threshold of the second detector is less than or equal to a second set value. In an embodiment of the present invention, the second setting value may be set to a relatively small value, for example, 0.3, and then the threshold of the second detector may be set to a value less than or equal to 0.3, so as to avoid missing the selected candidate QRS complexes.
Step S700: and calculating the similarity between the candidate QRS wave group and each group of characteristic templates, wherein the maximum value of the similarity is selected in the same group, and different parameter strategies are selected among different groups.
Specifically, within the same group, the result with the largest similarity value is selected; but because there are different groups, then different groups use different parameter strategies, for example, one of the groups uses a median, and the value of the similarity is calculated by the median; the other group uses the minimum value, and the value of the similarity is calculated through the minimum value, so that the optimal detection result is determined.
Step S800: and selecting a final result according to a set threshold value, and resampling to an original frequency. And finally, screening out a final result by using a threshold value for each candidate QRS complex, and resampling to the original frequency. That is, the result of each group is compared with a set threshold value, whether the result is greater than the set threshold value is determined, and then the final result is selected. Finally, the final result is resampled to the original frequency.
In a second aspect, in an embodiment of the present invention, there is provided an apparatus for detecting QRS complex based on feature template matching, as shown in fig. 2, the apparatus including:
a signal receiving unit configured to receive an electrocardiographic signal;
a preprocessing unit configured to preprocess the cardiac electrical signal, comprising: resampling the electrocardiosignals to a fixed frequency, and filtering;
a feature conversion unit configured to perform feature conversion on the preprocessed signal;
the first screening unit is used for acquiring a strict template QRS complex position by utilizing at least one first detector and adjusting a threshold value of the first detector; wherein the threshold of the base detector is greater than or equal to a first set value;
the template generating unit is configured to cut heartbeat feature segments of the preprocessed data and the feature-converted data by taking the QRS complex position of the template as a center, and divide different feature segments into a plurality of groups by using a similarity function to form a plurality of groups of feature templates;
a second screening unit configured to screen out candidate QRS complex positions by at least one second detector; wherein the threshold of the second detector is less than or equal to a second set value;
a calculating unit configured to calculate similarity between the candidate QRS complex and each set of feature templates; wherein, the maximum value of the similarity is selected in the same group, and different parameter strategies are selected among different groups;
and the third screening unit is configured to select a final result according to a set threshold value and resample the final result to the original frequency.
Specifically, the manner in which each unit implements its function corresponds to the corresponding step in the above method, and is not described here again.
In a third aspect, the present invention provides a storage medium having stored therein computer instructions, which when executed by a processor, are capable of performing any one of the methods for QRS complex detection based on feature template matching in the embodiments of the present invention. The specific implementation manner corresponds to the above description of the QRS complex detection method based on feature template matching, and details are not repeated here.
In embodiments of the invention, there may be any available media that can be accessed by a general purpose or special purpose computer. By way of example, the storage medium may include RAM, ROM, EPROM, E2PROM, registers, hard disk, removable disk, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other temporary or non-temporary medium that can be used to carry or store desired program code means in the form of instructions or data structures and that can be accessed by a general purpose or special purpose computer, or a general purpose or special purpose processor. Disk (disk) and disc (disc), as used herein, includes Compact Disc (CD), laser disc, optical disc, Digital Versatile Disc (DVD), floppy disk and blu-ray disc where disks usually reproduce data magnetically, while discs reproduce data optically with lasersThe data is replicated mathematically. Combinations of the above should also be included within the scope of computer-readable media. An exemplary storage medium is coupled to the processor such the processor can read information from, and write information to, the storage medium. In the alternative, the storage medium may be integral to the processor. The processor and the storage medium may reside in an ASIC. The ASIC may reside in a user terminal. In the alternative, the processor and the storage medium may reside as discrete components in a user terminal.
In a fourth aspect, the present invention provides an electronic device, as shown in fig. 3, which in an embodiment of the present invention includes any one of the storage media of the present invention and a processor, wherein the processor is configured to execute computer instructions stored in the storage medium. It should be noted that the electronic device may also include other components, such as an input device, a display device, etc., which are not shown for clarity of illustration of the principles of the present invention.
The above examples are only preferred embodiments of the present invention, it should be noted that: it will be apparent to those skilled in the art that various modifications and equivalents can be made without departing from the spirit of the invention, and it is intended that all such modifications and equivalents fall within the scope of the invention as defined in the claims.
Claims (10)
1. A QRS complex detection method based on feature template matching is characterized by comprising the following steps:
receiving an electrocardiosignal;
preprocessing the electrocardiosignals, comprising: resampling the electrocardiosignals to a fixed frequency, and filtering;
performing characteristic conversion on the preprocessed signals;
acquiring a strict position of a template QRS complex by utilizing at least one first detector and adjusting a threshold value of the first detector; wherein the threshold of the first detector is greater than or equal to a first set value;
cutting heartbeat feature fragments of the preprocessed data and the data after feature conversion by taking the QRS wave group position of the template as a center, and dividing different feature fragments into a plurality of groups by using a similarity function to form a plurality of groups of feature templates;
screening out candidate QRS complex positions through at least one second detector; wherein the threshold of the second detector is less than or equal to a second set value;
calculating the similarity between the candidate QRS wave group and each group of characteristic templates, wherein the maximum value of the similarity is selected in the same group as a result, and different parameter strategies are selected among different groups;
and selecting a final result according to a set threshold value, and resampling to an original frequency.
2. The method of claim 1, wherein the performing the feature transformation on the preprocessed signals comprises:
and performing feature transformation by adopting any one or any combination of more of derivatives, differences, Fourier transform, wavelet transform, Hilbert transform and information entropy.
3. The method of claim 2, wherein the first detector is any one or any combination of pantompkins1985, hamilton2002, martinez2003, christov2004, gamboa2008, elgendi2010, engzeemod2012, kallidas 2017, rodrigues2021, promac, wqrs, gqrs, xqrs open source detection algorithms.
4. The method of claim 3, wherein obtaining a rigorous template QRS complex position using at least one detector by adjusting a threshold of a first detector comprises:
respectively giving weight to each first detector, and calculating a first detection probability of a certain point according to whether the first detector detects the certain point and the weight corresponding to the first detector;
and comparing the first detection probability with a threshold value of a first detector, and determining the position of the QRS complex of the template according to the comparison result.
5. The method of claim 4, wherein the dividing the different feature segments into a plurality of groups using a similarity function, and wherein the forming the plurality of groups of feature templates comprises:
obtaining a standard template by performing weighted calculation on each dimension of all the fragments;
respectively calculating the similarity of each characteristic data segment and the standard template;
removing characteristic data segments with large difference with the identification degree of the standard template, sequencing the rest similarity, and dividing the characteristic data segments into a plurality of groups by taking a place with a high similarity change rate as a boundary of the groups;
and respectively calculating the weighted value of each group of characteristic data fragments on each dimension to obtain a characteristic data template.
6. The method of any one of claims 1 to 5, wherein the second detector is any one or any combination of pantompkins1985, hamilton2002, martinez2003, christov2004, gamboa2008, elgendi2010, engzeemod2012, kallidas 2017, rodrigues2021, promac, wqrs, gqrs, xqrs open source detection algorithms.
7. The method of claim 6, wherein the screening out the location of candidate QRS complexes by the second detector comprises:
respectively giving a weight to each second detector, and calculating a second detection probability of a certain point according to whether the second detector detects the certain point and the weight corresponding to the second detector;
and comparing the second detection probability with a threshold value of a basic detector, and determining the position of the candidate QRS complex according to the comparison result.
8. A QRS complex detection apparatus based on feature template matching, which detects QRS complexes according to the QRS complex detection method based on feature template matching of any one of claims 1 to 7, comprising:
a signal receiving unit configured to receive an electrocardiographic signal;
a preprocessing unit configured to preprocess the cardiac electrical signal, comprising: resampling the electrocardiosignals to a fixed frequency, and filtering;
a feature conversion unit configured to perform feature conversion on the preprocessed signal;
the first screening unit is used for acquiring a strict template QRS complex position by utilizing at least one first detector and adjusting a threshold value of the first detector; wherein the threshold of the base detector is greater than or equal to a first set value;
the template generating unit is configured to cut heartbeat feature segments of the preprocessed data and the feature-converted data by taking the QRS complex position of the template as a center, and divide different feature segments into a plurality of groups by using a similarity function to form a plurality of groups of feature templates;
a second screening unit configured to screen out candidate QRS complex positions by at least one second detector; wherein the threshold of the second detector is less than or equal to a second set value;
a calculating unit configured to calculate similarity between the candidate QRS complex and each set of feature templates; wherein, the maximum value of the similarity is selected in the same group, and different parameter strategies are selected among different groups;
and the third screening unit is configured to select a final result according to a set threshold value and resample the final result to the original frequency.
9. A storage medium, wherein the storage medium stores computer instructions, which when executed by a computer, can perform the method for QRS complex detection based on feature template matching according to any one of claims 1 to 7.
10. An electronic device, comprising:
the storage medium of claim 9; and the number of the first and second groups,
a processor capable of executing computer instructions stored in the storage medium.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111542679.3A CN114027853B (en) | 2021-12-16 | 2021-12-16 | QRS complex detection method, device, medium and equipment based on feature template matching |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111542679.3A CN114027853B (en) | 2021-12-16 | 2021-12-16 | QRS complex detection method, device, medium and equipment based on feature template matching |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114027853A true CN114027853A (en) | 2022-02-11 |
| CN114027853B CN114027853B (en) | 2022-09-27 |
Family
ID=80146907
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111542679.3A Active CN114027853B (en) | 2021-12-16 | 2021-12-16 | QRS complex detection method, device, medium and equipment based on feature template matching |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114027853B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115089196A (en) * | 2022-08-22 | 2022-09-23 | 博睿康科技（常州）股份有限公司 | Time phase detection method, time phase detection unit and closed-loop regulation and control system of online signal |
| 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 |
| WO2024061275A1 (en) * | 2022-09-20 | 2024-03-28 | 杭州辉图康科技有限公司 | Surface electrophysiological signal processing method for eliminating signal phase shift and system |
Citations (36)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0212370A2 (en) * | 1985-08-21 | 1987-03-04 | Kontron Instruments Holding N.V. | Respiration monitoring process and device |
| JPH0326233A (en) * | 1989-06-23 | 1991-02-04 | Fukuda Denshi Co Ltd | Method and device for detecting qrs |
| JPH0767844A (en) * | 1993-09-03 | 1995-03-14 | Fukuda Denshi Co Ltd | Device and method for detecting qrs wave in long time electrocardiogram |
| US20080194978A1 (en) * | 2004-05-01 | 2008-08-14 | Amir Beker | Apparatus and Method for Analysis of High Frequency Qrs Complexes |
| CN101856225A (en) * | 2010-06-30 | 2010-10-13 | 重庆大学 | Method for detecting R wave crest of electrocardiosignal |
| CN102835954A (en) * | 2012-09-07 | 2012-12-26 | 深圳市邦健电子有限公司 | Method for generating heart beating waveform template and treating ventricular premature beat detection and device |
| US20130237874A1 (en) * | 2012-03-12 | 2013-09-12 | Texas Instruments Incorporated | Real Time QRS Detection Using Adaptive Threshold |
| CN103549950A (en) * | 2013-11-19 | 2014-02-05 | 上海理工大学 | Improved difference threshold detection algorithm for mobile ECG (electrocardiogram) monitoring |
| CN103654770A (en) * | 2013-12-03 | 2014-03-26 | 上海理工大学 | Method and device for real-time detection of moving electrocardiosignal QRS (magnetic resonance angiography) waves |
| CN103714281A (en) * | 2013-12-12 | 2014-04-09 | 深圳先进技术研究院 | Identity recognition method based on electrocardiosignals |
| CN104055522A (en) * | 2014-07-01 | 2014-09-24 | 清华大学深圳研究生院 | Electrocardiosignal identity recognition method under arrhythmia condition |
| 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 |
| KR101524596B1 (en) * | 2014-02-28 | 2015-06-03 | 부산대학교 산학협력단 | PVC classification apparatus and method using QRS pattern, PVC pattern classification method and remote monitoring device |
| US20160278656A1 (en) * | 2013-12-20 | 2016-09-29 | Koninklijke Philips N.V. | Apparatus and method for determining the occurrance of a qrs complex in ecg data |
| CN106137184A (en) * | 2015-04-08 | 2016-11-23 | 四川锦江电子科技有限公司 | Electrocardiosignal QRS complex detection method based on wavelet transformation |
| US20170105643A1 (en) * | 2015-06-23 | 2017-04-20 | Shenzhen Institutes Of Advanced Technology Chinese Academy Of Sciences | Gpu-based parallel electrocardiogram signal analysis method, computer readable storage medium and device |
| CN106963368A (en) * | 2017-02-07 | 2017-07-21 | 深圳中科汇康技术有限公司 | QRS complex detection method and device |
| CN107495959A (en) * | 2017-07-27 | 2017-12-22 | 大连大学 | A kind of electrocardiosignal sorting technique based on one-dimensional convolutional neural networks |
| CN107788969A (en) * | 2017-09-29 | 2018-03-13 | 成都瑞迪康医疗科技有限公司 | The automatic testing method of QRS complex in a kind of electrocardiosignal |
| CN108056773A (en) * | 2017-12-11 | 2018-05-22 | 重庆邮电大学 | Based on the algorithms of QRS complexes detection in electrocardiogram signal for improving variation mode decomposition |
| CN108509823A (en) * | 2017-02-24 | 2018-09-07 | 深圳市理邦精密仪器股份有限公司 | The detection method and device of QRS complex |
| CN109009084A (en) * | 2018-06-08 | 2018-12-18 | 广州视源电子科技股份有限公司 | QRS complex method of calibration, device and the equipment of multi-lead electrocardiosignal, medium |
| CN109330584A (en) * | 2018-11-08 | 2019-02-15 | 山东大学 | Electrocardiosignal personal identification method and system based on dictionary learning and rarefaction representation |
| CN109620214A (en) * | 2018-12-07 | 2019-04-16 | 上海数创医疗科技有限公司 | ST sections of automatic judging methods of electrocardiosignal and device based on artificial intelligence technology |
| CN110123308A (en) * | 2019-05-28 | 2019-08-16 | 哈尔滨理工大学 | The detection method of noninvasive Fetal ECG signal R wave |
| CN110367973A (en) * | 2019-07-01 | 2019-10-25 | 复旦大学 | Multi-lead electrocardiosignal QRS complex is all risen the detection method of initial point |
| CN110929759A (en) * | 2019-10-24 | 2020-03-27 | 东软集团股份有限公司 | Training device and method for detection model and electrocardio data processing method and device |
| CN111166322A (en) * | 2020-01-17 | 2020-05-19 | 哈尔滨工业大学 | Electrocardiosignal characteristic wave extraction method |
| CN111345816A (en) * | 2020-02-25 | 2020-06-30 | 广州视源电子科技股份有限公司 | Multi-lead QRS complex detection method, device, equipment and storage medium |
| CN111345815A (en) * | 2020-02-11 | 2020-06-30 | 广州视源电子科技股份有限公司 | Method, device, equipment and storage medium for detecting QRS wave in electrocardiosignal |
| US20210052180A1 (en) * | 2019-08-22 | 2021-02-25 | West Affum Holdings Corp. | Cardiac Monitoring System with Normally Conducted QRS Complex Identification |
| CN112438735A (en) * | 2021-01-29 | 2021-03-05 | 深圳市理邦精密仪器股份有限公司 | Electrocardiogram P wave detection method, analysis device and storage medium |
| KR20210044163A (en) * | 2019-10-14 | 2021-04-22 | 고려대학교 세종산학협력단 | Prediction of Ventricular Fibrillation by Machine Learning Based on QRS Complex Shape Features |
| CN113191249A (en) * | 2021-04-28 | 2021-07-30 | 深圳邦健生物医疗设备股份有限公司 | Electrocardiosignal template matching method, device, equipment and medium |
| US11077310B1 (en) * | 2016-10-04 | 2021-08-03 | West Affum Holdings Corp. | Wearable cardioverter defibrillator (WCD) system detecting QRS complexes in ECG signal by matched difference filter |
| CN113197584A (en) * | 2021-04-30 | 2021-08-03 | 扬州大学 | QRS wave group identification method based on difference zero-crossing detection method |
-
2021
- 2021-12-16 CN CN202111542679.3A patent/CN114027853B/en active Active
Patent Citations (36)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0212370A2 (en) * | 1985-08-21 | 1987-03-04 | Kontron Instruments Holding N.V. | Respiration monitoring process and device |
| JPH0326233A (en) * | 1989-06-23 | 1991-02-04 | Fukuda Denshi Co Ltd | Method and device for detecting qrs |
| JPH0767844A (en) * | 1993-09-03 | 1995-03-14 | Fukuda Denshi Co Ltd | Device and method for detecting qrs wave in long time electrocardiogram |
| US20080194978A1 (en) * | 2004-05-01 | 2008-08-14 | Amir Beker | Apparatus and Method for Analysis of High Frequency Qrs Complexes |
| CN101856225A (en) * | 2010-06-30 | 2010-10-13 | 重庆大学 | Method for detecting R wave crest of electrocardiosignal |
| US20130237874A1 (en) * | 2012-03-12 | 2013-09-12 | Texas Instruments Incorporated | Real Time QRS Detection Using Adaptive Threshold |
| CN102835954A (en) * | 2012-09-07 | 2012-12-26 | 深圳市邦健电子有限公司 | Method for generating heart beating waveform template and treating ventricular premature beat detection and device |
| CN103549950A (en) * | 2013-11-19 | 2014-02-05 | 上海理工大学 | Improved difference threshold detection algorithm for mobile ECG (electrocardiogram) monitoring |
| CN103654770A (en) * | 2013-12-03 | 2014-03-26 | 上海理工大学 | Method and device for real-time detection of moving electrocardiosignal QRS (magnetic resonance angiography) waves |
| CN103714281A (en) * | 2013-12-12 | 2014-04-09 | 深圳先进技术研究院 | Identity recognition method based on electrocardiosignals |
| US20160278656A1 (en) * | 2013-12-20 | 2016-09-29 | Koninklijke Philips N.V. | Apparatus and method for determining the occurrance of a qrs complex in ecg data |
| KR101524596B1 (en) * | 2014-02-28 | 2015-06-03 | 부산대학교 산학협력단 | PVC classification apparatus and method using QRS pattern, PVC pattern classification method and remote monitoring device |
| CN104055522A (en) * | 2014-07-01 | 2014-09-24 | 清华大学深圳研究生院 | Electrocardiosignal identity recognition method under arrhythmia condition |
| 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 |
| CN106137184A (en) * | 2015-04-08 | 2016-11-23 | 四川锦江电子科技有限公司 | Electrocardiosignal QRS complex detection method based on wavelet transformation |
| US20170105643A1 (en) * | 2015-06-23 | 2017-04-20 | Shenzhen Institutes Of Advanced Technology Chinese Academy Of Sciences | Gpu-based parallel electrocardiogram signal analysis method, computer readable storage medium and device |
| US11077310B1 (en) * | 2016-10-04 | 2021-08-03 | West Affum Holdings Corp. | Wearable cardioverter defibrillator (WCD) system detecting QRS complexes in ECG signal by matched difference filter |
| CN106963368A (en) * | 2017-02-07 | 2017-07-21 | 深圳中科汇康技术有限公司 | QRS complex detection method and device |
| CN108509823A (en) * | 2017-02-24 | 2018-09-07 | 深圳市理邦精密仪器股份有限公司 | The detection method and device of QRS complex |
| CN107495959A (en) * | 2017-07-27 | 2017-12-22 | 大连大学 | A kind of electrocardiosignal sorting technique based on one-dimensional convolutional neural networks |
| CN107788969A (en) * | 2017-09-29 | 2018-03-13 | 成都瑞迪康医疗科技有限公司 | The automatic testing method of QRS complex in a kind of electrocardiosignal |
| CN108056773A (en) * | 2017-12-11 | 2018-05-22 | 重庆邮电大学 | Based on the algorithms of QRS complexes detection in electrocardiogram signal for improving variation mode decomposition |
| CN109009084A (en) * | 2018-06-08 | 2018-12-18 | 广州视源电子科技股份有限公司 | QRS complex method of calibration, device and the equipment of multi-lead electrocardiosignal, medium |
| CN109330584A (en) * | 2018-11-08 | 2019-02-15 | 山东大学 | Electrocardiosignal personal identification method and system based on dictionary learning and rarefaction representation |
| CN109620214A (en) * | 2018-12-07 | 2019-04-16 | 上海数创医疗科技有限公司 | ST sections of automatic judging methods of electrocardiosignal and device based on artificial intelligence technology |
| CN110123308A (en) * | 2019-05-28 | 2019-08-16 | 哈尔滨理工大学 | The detection method of noninvasive Fetal ECG signal R wave |
| CN110367973A (en) * | 2019-07-01 | 2019-10-25 | 复旦大学 | Multi-lead electrocardiosignal QRS complex is all risen the detection method of initial point |
| US20210052180A1 (en) * | 2019-08-22 | 2021-02-25 | West Affum Holdings Corp. | Cardiac Monitoring System with Normally Conducted QRS Complex Identification |
| KR20210044163A (en) * | 2019-10-14 | 2021-04-22 | 고려대학교 세종산학협력단 | Prediction of Ventricular Fibrillation by Machine Learning Based on QRS Complex Shape Features |
| CN110929759A (en) * | 2019-10-24 | 2020-03-27 | 东软集团股份有限公司 | Training device and method for detection model and electrocardio data processing method and device |
| CN111166322A (en) * | 2020-01-17 | 2020-05-19 | 哈尔滨工业大学 | Electrocardiosignal characteristic wave extraction method |
| CN111345815A (en) * | 2020-02-11 | 2020-06-30 | 广州视源电子科技股份有限公司 | Method, device, equipment and storage medium for detecting QRS wave in electrocardiosignal |
| CN111345816A (en) * | 2020-02-25 | 2020-06-30 | 广州视源电子科技股份有限公司 | Multi-lead QRS complex detection method, device, equipment and storage medium |
| CN112438735A (en) * | 2021-01-29 | 2021-03-05 | 深圳市理邦精密仪器股份有限公司 | Electrocardiogram P wave detection method, analysis device and storage medium |
| CN113191249A (en) * | 2021-04-28 | 2021-07-30 | 深圳邦健生物医疗设备股份有限公司 | Electrocardiosignal template matching method, device, equipment and medium |
| CN113197584A (en) * | 2021-04-30 | 2021-08-03 | 扬州大学 | QRS wave group identification method based on difference zero-crossing detection method |
Non-Patent Citations (4)
| Title |
|---|
| 刘雄飞等: "基于提升小波变换和多种策略的QRS波检测算法", 《中国医学物理学杂志》 * |
| 李鹏等: "一种QRS波群实时检测方法", 《生物物理学报》 * |
| 王利琴等: "一种实时鲁棒的QRS波群检测算法", 《计算机应用与软件》 * |
| 苏丽等: "心电信号QRS波群检测算法研究", 《哈尔滨工程大学学报》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115089196A (en) * | 2022-08-22 | 2022-09-23 | 博睿康科技（常州）股份有限公司 | Time phase detection method, time phase detection unit and closed-loop regulation and control system of online signal |
| CN115089196B (en) * | 2022-08-22 | 2022-11-11 | 博睿康科技（常州）股份有限公司 | Time phase detection method, time phase detection unit and closed-loop regulation and control system of online signal |
| WO2024061275A1 (en) * | 2022-09-20 | 2024-03-28 | 杭州辉图康科技有限公司 | Surface electrophysiological signal processing method for eliminating signal phase shift and system |
| 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 |
|---|---|
| CN114027853B (en) | 2022-09-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114027853B (en) | QRS complex detection method, device, medium and equipment based on feature template matching | |
| Wang et al. | Phonocardiographic signal analysis method using a modified hidden Markov model | |
| Karar et al. | Automated diagnosis of heart sounds using rule-based classification tree | |
| Papadaniil et al. | Efficient heart sound segmentation and extraction using ensemble empirical mode decomposition and kurtosis features | |
| Yuenyong et al. | A framework for automatic heart sound analysis without segmentation | |
| Dokur et al. | Heart sound classification using wavelet transform and incremental self-organizing map | |
| Tang et al. | PCG classification using multidomain features and SVM classifier | |
| Gavrovska et al. | Classification of prolapsed mitral valve versus healthy heart from phonocardiograms by multifractal analysis | |
| Daqrouq et al. | Neural network and wavelet average framing percentage energy for atrial fibrillation classification | |
| CN112971839B (en) | Heart sound classification method based on feedforward convolution neural network | |
| Javed et al. | A signal processing module for the analysis of heart sounds and heart murmurs | |
| Grønnesby et al. | Feature extraction for machine learning based crackle detection in lung sounds from a health survey | |
| Gavrovska et al. | Automatic heart sound detection in pediatric patients without electrocardiogram reference via pseudo-affine Wigner–Ville distribution and Haar wavelet lifting | |
| Sedighian et al. | Pediatric heart sound segmentation using Hidden Markov Model | |
| Arpitha et al. | Spectrogram analysis of ECG signal and classification efficiency using MFCC feature extraction technique | |
| CN112971795B (en) | Electrocardiosignal quality evaluation method | |
| Pretorius et al. | Development of a pediatric cardiac computer aided auscultation decision support system | |
| CN111368627A (en) | Heart sound classification method and system based on CNN combined with improved frequency wavelet slice transformation | |
| Singh et al. | Short unsegmented PCG classification based on ensemble classifier | |
| Omarov et al. | Artificial Intelligence in Medicine: Real Time Electronic Stethoscope for Heart Diseases Detection. | |
| Bourouhou et al. | Heart sound signals segmentation and multiclass classification | |
| Mustafa et al. | Detection of heartbeat sounds arrhythmia using automatic spectral methods and cardiac auscultatory | |
| Wan et al. | A novel atrial fibrillation automatic detection algorithm based on ensemble learning and multi-feature discrimination | |
| Bourouhou et al. | Heart sounds classification for a medical diagnostic assistance | |
| Sofwan et al. | Normal and Murmur Heart Sound Classification Using Linear Predictive Coding and k-Nearest Neighbor Methods |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |