CN109288515B - Method and device for periodic monitoring of premature beats in wearable ECG signals - Google Patents
Method and device for periodic monitoring of premature beats in wearable ECG signals Download PDFInfo
- Publication number
- CN109288515B CN109288515B CN201811350454.6A CN201811350454A CN109288515B CN 109288515 B CN109288515 B CN 109288515B CN 201811350454 A CN201811350454 A CN 201811350454A CN 109288515 B CN109288515 B CN 109288515B
- Authority
- CN
- China
- Prior art keywords
- beat
- premature
- template
- heart
- beats
- 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.)
- Active
Links
- 208000000418 Premature Cardiac Complexes Diseases 0.000 title claims abstract description 316
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000012544 monitoring process Methods 0.000 title claims abstract description 32
- 206010015856 Extrasystoles Diseases 0.000 title claims description 25
- 230000000737 periodic effect Effects 0.000 title claims description 11
- 238000001514 detection method Methods 0.000 claims abstract description 194
- 230000002861 ventricular Effects 0.000 claims description 102
- 206010042602 Supraventricular extrasystoles Diseases 0.000 claims description 13
- 238000004891 communication Methods 0.000 claims description 13
- 206010003658 Atrial Fibrillation Diseases 0.000 claims description 10
- 238000012806 monitoring device Methods 0.000 claims description 7
- 239000000284 extract Substances 0.000 claims description 6
- 230000007774 longterm Effects 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 5
- 238000011897 real-time detection Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 2
- 208000009729 Ventricular Premature Complexes Diseases 0.000 claims 16
- 208000002102 Atrial Premature Complexes Diseases 0.000 claims 2
- 230000000694 effects Effects 0.000 description 6
- 230000006872 improvement Effects 0.000 description 6
- 230000016507 interphase Effects 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 230000002028 premature Effects 0.000 description 4
- 238000002360 preparation method Methods 0.000 description 4
- 238000012216 screening Methods 0.000 description 4
- 238000003745 diagnosis Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 241001347978 Major minor Species 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 2
- 206010003119 arrhythmia Diseases 0.000 description 2
- 230000006793 arrhythmia Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 208000019622 heart disease Diseases 0.000 description 2
- 238000003780 insertion Methods 0.000 description 2
- 230000037431 insertion Effects 0.000 description 2
- 230000000877 morphologic effect Effects 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 208000009525 Myocarditis Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 229940127554 medical product Drugs 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 208000004124 rheumatic heart disease Diseases 0.000 description 1
- 208000003663 ventricular fibrillation Diseases 0.000 description 1
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]
-
- 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
- A61B5/7271—Specific aspects of physiological measurement analysis
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Biophysics (AREA)
- General Health & Medical Sciences (AREA)
- Veterinary Medicine (AREA)
- Public Health (AREA)
- Animal Behavior & Ethology (AREA)
- Surgery (AREA)
- Molecular Biology (AREA)
- Medical Informatics (AREA)
- Artificial Intelligence (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Signal Processing (AREA)
- Psychiatry (AREA)
- Physiology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Fuzzy Systems (AREA)
- Cardiology (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The invention discloses a periodicity monitoring method and device based on a premature beat signal in a wearable electrocardiosignal, which comprises a wearable electrocardiosignal detection module and a long-time premature beat detection module, wherein an electrocardiosignal reading unit in the long-time premature beat detection module reads a heart beat to be detected, after the heart beat type is determined by a real-time premature beat detection unit, a primary template and a secondary template are respectively generated by a primary template generation unit and a secondary template generation unit, and then the primary and secondary heart beat detection unit finally determines the heart beat type.
Description
Field of the invention
The invention relates to the technical field of signal detection and medical equipment electronics, in particular to a periodicity monitoring method and device based on a premature beat signal in a wearable electrocardiosignal.
Background
Premature beats, also known as extra-systoles or extra-systoles, are the most common and frequent arrhythmias. Depending on the mechanism of occurrence, it can be divided into two types, ventricular premature beats and supraventricular premature beats, the most common of which are ventricular premature beats and the second of which are supraventricular premature beats. Premature beat can occur in normal people and patients with organic heart diseases, is commonly seen in myocarditis, rheumatic heart disease and the like, correctly detects a premature beat signal, is a key for improving the detection accuracy of arrhythmia events, and has important practical value for improving heart disease diagnosis and monitoring serious patients.
As is well known, for the control and prevention of such diseases, the electrocardiographic data needs to be recorded and monitored in real time for a long time, and good control effect can be achieved through early analysis and intervention. If the patient chooses to go to a hospital for a long-term examination, the hospitalizing process is complicated, the cost is high, and the patient cannot bear the medical examination in ordinary families. The existing household electrocardio monitor in the market still has the problems of large volume and incapability of real-time diagnosis, and a wearable electrocardio monitoring system not only saves the cost of equipment and reduces the volume of the equipment, but also can realize local real-time electrocardio analysis and remote information communication and becomes the main trend of mobile medical products. Meanwhile, the computing power of the mobile platform is limited, so that the time complexity and the space complexity of an analysis algorithm running on the platform are not too large.
The current automatic premature beat diagnosis algorithm comprises a neural network method, a support vector machine, deep learning and the like, wherein the methods mainly extract time-frequency characteristics of the electrocardiosignal and then input the characteristics into the neural network or the support vector machine for classification. Under the condition, the process of feature extraction is complex, the automatic electrocardiogram analysis is obviously influenced by the feature extraction result, and more feature sets are generally needed to realize a better classification effect, so that the operation complexity of the terminal is increased in the process of remote transmission, the mobile platform is difficult to load, and the monitoring effect cannot be effectively realized; on the other hand, in order to realize the real-time performance of premature beat monitoring, most of the current premature beat detection algorithms are based on a short-time data segment, the data information contained in the short-time data segment is less, especially dynamic electrocardio signals collected in a wearing environment have large baseline drift and much motion noise, and the premature beat detection on the short-time data segment is easy to cause false detection and missed detection, so that a periodic and long-time wearable detection device and algorithm need to be developed to realize comfortable, simple and quick premature beat monitoring.
Disclosure of Invention
The invention provides a periodic monitoring method and device based on a premature beat signal in wearable electrocardiosignals, which aims at the problems in the prior art, and comprises a wearable electrocardiosignal detection module and a long-time premature beat detection module, wherein an electrocardiosignal reading unit in the long-time premature beat detection module reads a heart beat to be detected, after the heart beat type is determined by the real-time premature beat detection unit, a primary template generation unit and a secondary template generation unit respectively generate a primary template and a secondary template, then the primary template and the secondary template generation unit finally determine the heart beat type, and on the basis of normal heart beat, the ventricular premature beat detection unit judges whether the heart beat is a supraventricular premature beat or a normal heart beat, so that the long-time comfortable monitoring of ventricular premature beat and supraventricular premature beat under daily activities of people is realized, the algorithm is simple, the operation is convenient, and the device is suitable for mobile medical treatment, The fields of sports monitoring and medical monitoring, etc.
In order to achieve the purpose, the invention adopts the technical scheme that: the periodicity monitoring method based on the premature beat signal in the wearable electrocardiosignal comprises the following steps:
s1, reading electrocardiosignals: the electrocardiosignal is extracted once every 5 minutes, and the signal extracted within 5 minutes is divided into a plurality of 10-second electrocardio data in a mode of overlapping 2 seconds;
s2, detecting premature beat in real time: the steps are that the characteristic parameters of the heart beat to be detected are extracted, and parameter calculation and comparison are sequentially carried out on the characteristic parameters and the detection of the quasi-premature beat heart beat, the detection of the ventricular premature beat, the detection of the supraventricular premature beat and the verification of the premature beat heart beat, so that the monitoring and classification of the electrocardiosignals are realized;
s3, generating a main class template: determining a main class template according to the premature beat detection result after the real-time premature beat detection in the step S2 and the correlation coefficient among heartbeats in the 5-minute data;
s4, generating a secondary template: determining a secondary template according to the premature beat detection result after the real-time premature beat detection in the step S2 and the correlation coefficient among heartbeats in the 5-minute data;
s5, primary and secondary heartbeat detection: determining the type of each heart beat according to the existence of the secondary template and the correlation coefficient between each heart beat and the primary and secondary templates;
s6, supraventricular premature beat detection: and judging the supraventricular premature beat heart beat according to the characteristic parameters of the normal heart beat and the atrial fibrillation detection result determined in the step S5.
As a refinement of the present invention, the step S3 further includes:
s31, extracting the detection results R1 of the ventricular premature beat and the supraventricular premature beat obtained after the real-time premature beat detection in the step S2;
s32, calculating the number of heartbeats of each type in R1, and generating a heartbeat P1 by taking the heartbeat with the largest number as a preparation main type template;
s33, calculating the correlation coefficient C1 among the heart beats in the preliminary main class template generation heart beat P1;
s34, sorting C1 in descending order, and selecting the heart beats ranked at the top 1/3 as a quasi-main template to generate heart beats P2;
s35, calculating the correlation coefficient C2 among the heart beats in the quasi-master template generation heart beat P2;
s36, sorting C2 in descending order, selecting the heart beats ranked at the top 1/3 as a main template to generate heart beats, and constructing a main template T1 according to the heart beats;
s37, calculating QRS width of the main template T1, and judging the type of the main template.
As a refinement of the present invention, the step S4 further includes:
s41, extracting the detection results R1 of the ventricular premature beat and the supraventricular premature beat obtained after the real-time premature beat detection in the step S2;
s42, calculating the number of heartbeats of each type in R1, and generating a heart beat P3 by using the heart beat with the second most number as a preparation secondary template;
s43, judging whether the number of the heart beats P3 generated by the preliminary class template is more than 10; if yes, go to step S44; if not, the output secondary template does not exist;
s44, calculating the correlation coefficient C3 among each heartbeat in the preliminary secondary template generation heartbeat P3;
s45, sorting C3 in descending order, and selecting the heart beats ranked at the top 1/3 as a quasi-class template to generate heart beats P4;
s46, calculating a correlation coefficient C4 among the heartbeats in the quasi-class template generation heartbeat P4;
s47, sorting C4 in descending order, selecting the heart beats ranked at the top 1/3 as a secondary template to generate heart beats, and constructing a primary template T2 according to the heart beats;
s48, determining the type of the minor class template T2 according to the type of the major class template T1.
As still another improvement of the present invention, the step S5 further includes:
s51, extracting a main template T1 and a minor template T2;
s52, judging whether a minor template T2 exists; if so, continue to step S53; if not, go to step S57;
s53, calculating a correlation coefficient C5 between each heartbeat and the primary template and a correlation coefficient C6 between each heartbeat and the secondary template;
s54, judging whether the correlation coefficient C5 between the main type templates is more than 0.9; if so, continue to step S55; if not, go to step S56;
s55, judging whether the correlation coefficient C6 between the secondary templates is more than 0.9; if so, outputting noise in the current heartbeat type; if the result is negative, outputting the current heartbeat type as a main heartbeat type;
s56, judging whether the correlation coefficient C6 between the secondary templates is more than 0.9; if yes, outputting the current heartbeat type as a secondary heartbeat type; if the result is negative, outputting noise in the current heartbeat type;
s57, calculating a correlation coefficient C7 between each heartbeat and the main type template;
s58, judging whether the correlation coefficient C7 between the main type templates is more than 0.9; if so, outputting the current heartbeat type as a main heartbeat type; if not, the current heartbeat type is output as a secondary heartbeat type.
As still another improvement of the present invention, the step S5 further includes:
s61, extracting the RR interval mean values of the normal heart beat detected in the step S4, the RR interval mean values of the 5-minute electrocardiosignals and the RR interval mean values of the normal heart beat detected in the step S4;
s62, judging whether the RR interval before the normal heartbeat is less than the RR interval mean value of the electrocardiosignal of 5 minutes, whether the RR interval after the normal heartbeat is more than the RR interval mean value of the electrocardiosignal of 5 minutes, and whether the sum of the RR interval before the normal heartbeat and the RR interval after the normal heartbeat is less than or equal to twice the RR interval mean value of the electrocardiosignal of 5 minutes; if yes, go to step S63; if the result is negative, the current heart beat output is the normal heart beat;
s63, judging whether the 30-second electrocardiogram data of the current heart beat is an atrial fibrillation signal; if so, outputting the current heart beat as a normal heart beat; if not, the current heartbeat output is the supraventricular premature beat.
As another improvement of the present invention, the step S2 further includes:
s21, preprocessing the electrocardiosignal;
s22, extracting electrocardiosignal characteristic parameters: extracting characteristic parameters of the heart beat to be detected from the electrocardiosignals preprocessed in the step S21, wherein the characteristic parameters comprise the R wave position of the heart beat to be detected in the 10-second electrocardiosignals, the RR intervals of all the heart beats to be detected, the RR interval mean value of the 10-second electrocardiosignals, the R wave amplitude of all the heart beats to be detected, the R wave amplitude mean value of the 10-second electrocardiosignals, the QRS wave width of all the heart beats to be detected, the QRS wave width mean value of the 10-second electrocardiosignals, the R wave correlation coefficients of all the heart beats to be detected and the R wave correlation number mean value of the 10-second electrocardiosignals;
s23, judging the premature beat: judging whether the electrocardiosignal is a quasi-premature beat heartbeat according to the RR interval of the heartbeat to be detected, if so, continuing to step S24; if not, the signal is normal heartbeat;
s24, judging ventricular premature beat: judging whether the quasi-premature beat screened in the step S23 is a ventricular premature beat or not according to the R wave correlation coefficient, the QRS wave width and the R wave amplitude of the beat to be measured, if not, screening out a non-ventricular premature beat, and continuing to the step S25; if yes, the signal is ventricular premature beat;
s25, judging the supraventricular premature beat: judging whether the ventricular premature beat is the supraventricular premature beat or not according to the R wave correlation coefficient of the non-ventricular premature beat screened in the step S24 and the R wave correlation number average value of the 10-second electrocardiosignal, and if so, continuing to perform the step S26; if not, the signal is normal heartbeat;
s26, checking the premature beat and the heart beat: the premature beat check comprises supraventricular premature beat self-check, normal ventricular premature beat self-check and ventricular premature beat mutual-check, and is used for judging whether the supraventricular premature beat has a false-check ventricular premature beat, whether the normal ventricular premature beat has a false-check noise signal and whether the normal beat has a missed-check ventricular premature beat.
As a further improvement of the present invention, the ventricular premature beat in step S24 includes a normal ventricular premature beat, an insertion ventricular premature beat, and a continuous ventricular premature beat, and the method for determining the continuous premature beat includes: taking R wave correlation coefficients of adjacent normal beats of the normal ventricular premature beat heart beat, and judging whether the correlation coefficients are smaller than the R wave correlation number mean value of the 10-second electrocardio data and whether the correlation coefficients are between 80% and 120% of the R wave correlation number mean value of the 10-second electrocardio signal; if yes, outputting the current normal ventricular premature beat adjacent to the heart beat and the current normal ventricular premature beat as a continuous ventricular premature beat.
In order to achieve the purpose, the invention adopts the technical scheme that: the periodicity monitoring device based on the premature beat signal in the wearable electrocardiosignal comprises a wearable electrocardiosignal detection module and a long-time premature beat detection module,
the wearable electrocardiosignal detection module comprises a dry electrode unit for signal sensing, a signal detection unit for signal processing and a communication module unit for real-time communication;
the long-time premature beat detection module comprises an electrocardiosignal reading unit for reading signals, a real-time premature beat detection unit for detecting a premature beat in real time, a main template generation unit for generating a main heartbeat template, a secondary template generation unit for generating a secondary heartbeat template, a main and secondary heartbeat detection unit for detecting the main and secondary heartbeats and an supraventricular premature beat detection unit for detecting the supraventricular premature beat;
the electrocardiosignal reading unit reads the heart beat to be detected, after the heart beat type is determined by the real-time premature beat detection unit, the primary template generation unit and the secondary template generation unit respectively generate a primary template and a secondary template, then the primary and secondary heart beat detection unit finally determines the heart beat type, and on the basis of normal heart beat, the ventricular premature beat detection unit judges whether the heart beat is a ventricular premature beat or a normal heart beat.
As an improvement of the invention, the wearable electrocardiosignal detection module is connected with the long-time premature beat detection module in a wired or wireless way, and the long-time premature beat monitoring module is positioned in an internal system of the wearable electrocardiosignal detection module; or the wearable electrocardiosignal monitoring module is separated and is positioned in an external mobile platform.
As a further improvement of the invention, the number and the layout of the dry electrode units are not unique, and the monitoring devices forming different leads
Compared with the prior art, the invention has the beneficial effects that:
1. the adopted wearable electrocardiosignal detection module can adjust the number and the arrangement positions of the dry electrodes according to specific monitoring requirements to form a wearable electrocardiosignal detection device with different leads, so that the adaptability of the device is improved, and the application range of the device is widened;
2. the adopted real-time premature beat detection module only needs to simply preprocess an algorithm, extracts simple characteristic parameters such as an R wave position, an RR interphase, an average RR interphase, an R wave amplitude, a QRS wave width, an R wave correlation coefficient and the like, and then judges ventricular premature beats and supraventricular premature beats by integrating the characteristic parameters, so that the ventricular premature beats and the supraventricular premature beats can be detected more simply, quickly and accurately compared with the prior art, the real-time performance and the accuracy rate of the algorithm are improved, the calculated amount of the algorithm is reduced, and the method can be quickly transplanted to different mobile terminals for relevant application of mobile medical treatment;
3. the long-time premature beat detection module is adopted, only single-lead electrocardiosignals are processed, so that ventricular premature beat and supraventricular premature beat can be accurately monitored for a long time, redundant errors of multi-lead electrocardiosignals cannot be introduced in the application of dynamic electrocardio data, and the detection efficiency is high;
4. the long-time premature beat heart beat detection module is only used for simple characteristic parameters of the QRS waves such as RR intervals, R wave amplitude values, QRS wave width, R wave correlation coefficients and the like, does not relate to too many morphological characteristics of the electrocardiographic wave, and has good processing effect and high detection accuracy for dynamic electrocardiographic data with large baseline drift and much motion noise;
5. the long-time premature beat detection module is adopted to detect ventricular premature beat and supraventricular premature beat according to the specific characteristics of different subtypes, the detection accuracy is high, and meanwhile, according to the detection results of the ventricular premature beat and the supraventricular premature beat, the subtype classification (such as bigeminal rule, ventricular velocity and ventricular fibrillation) can be further carried out only by simple operation, and the detection is convenient and rapid.
6. The wearable electrocardiosignal detection module is connected with the long-time premature beat detection module in a wired or wireless mode, the long-time premature beat detection module can be positioned in an internal system of the wearable electrocardiosignal detection module, is integrated, can be separated from the wearable electrocardiosignal monitoring module and is positioned in an external mobile platform, the wearable electrocardiosignal detection module is simple in structure, convenient and fast to operate, flexible and changeable, suitable for requirements of different users, and capable of achieving real-time and light-weight effective monitoring of premature beat signals of the wearable electrocardiosignals;
7. the long-time premature beat detection module is adopted, and an atrial fibrillation detection link is introduced, so that the misjudgment of the supraventricular premature beat is reduced, and the detection accuracy of the supraventricular premature beat is increased;
8. the long-time premature beat detection module is adopted to generate the primary template and the secondary template on the basis of a real-time detection result, so that the accuracy of the templates is high, and the detection accuracy of the ventricular premature beat and the supraventricular premature beat is further improved;
9. the adopted primary template generation unit and the secondary template generation unit adopt a method of multiple matching screening to generate the templates, so that the accuracy of the templates is improved;
10. the main and secondary template generation units are adopted, on the basis of the real-time detection result, accurate template generation can be realized only through one characteristic parameter of the correlation coefficient, and the algorithm is simple, quick and accurate.
Drawings
FIG. 1 is a schematic structural diagram of a wearable device for monitoring the periodicity of a premature beat signal in an electrocardiographic signal according to the present invention;
FIG. 2 is a schematic structural diagram of a 12-lead split monitoring device in embodiment 2 of the present invention;
FIG. 3 is a schematic flow chart of the real-time premature beat detection in step S2 according to the present invention;
FIG. 4 is a schematic flow chart of the step S23 of determining the quasi-premature beat according to the present invention;
FIG. 5 is a schematic flow chart illustrating the step S24 of determining ventricular premature beat according to the present invention;
FIG. 6 is a schematic flow chart illustrating the step S25 of determining the ventricular premature beat;
FIG. 7 is a flowchart illustrating a method for verifying a self-check link of a supraventricular premature beat in the step S26 according to the present invention;
FIG. 8 is a flowchart illustrating a method for verifying a self-check link of ventricular premature beats in the step S26 according to the present invention;
FIG. 9 is a flowchart illustrating a method for performing a ventricular premature beat mutual check in the step S26;
FIG. 10 is a schematic flow chart of the main class template generating unit in the long premature beat detection module according to the present invention;
FIG. 11 is a schematic flow chart of the secondary template generating unit in the long premature beat detection module according to the present invention;
FIG. 12 is a schematic flow chart of the primary and secondary heartbeat detection units in the long premature beat detection module according to the present invention;
FIG. 13 is a schematic flow chart of the supraventricular premature beat detection unit in the long term premature beat detection module according to the present invention.
Reference numerals: 1. the wearable electrocardiosignal detection module comprises 110 dry electrode units, 130 signal detection units, 150 communication module units;
2. the long time premature beat detection module comprises a long time premature beat detection module, 210, an electrocardiosignal reading unit, 220, a real time premature beat detection unit, 230, a main class template generation unit, 240, a secondary class template generation unit, 250, a main class and secondary class heartbeat detection unit, 260 and a supraventricular premature beat detection unit.
Detailed Description
The invention will be explained in more detail below with reference to the drawings and examples.
Example 1
A periodicity monitoring device based on a premature beat signal in a wearable electrocardiosignal is shown in figure 1 and comprises a wearable electrocardiosignal detection module 1 and a long-time premature beat detection module 2,
the wearable electrocardiosignal detection module 1 comprises a dry electrode unit 110 for signal sensing, a signal detection unit 130 for signal processing and a communication module unit 150 for real-time communication, the dry electrode unit 110 can continuously acquire electrocardio data of a human body in real time on the premise of ensuring comfort, one end of the signal detection unit 130 is connected with the dry electrode unit 110, and the other end of the signal detection unit 130 is connected with the communication module unit 150; the signal detection unit 130 converts the analog electrocardiosignals detected by the dry electrode unit 110 into digital signals, and sends the digital signals to the long-time premature beat detection module 2 through the communication module unit 150; the communication module unit 150 communicates with the long premature beat detection module 2 in real time in a wired or wireless manner;
the long-time premature beat detection module 2 comprises an electrocardiosignal reading unit 210 for reading signals, a real-time premature beat detection unit 220 for detecting a premature beat in real time, a main class template generation unit 230 for generating a main class heartbeat template, a secondary class template generation unit 240 for generating a secondary class heartbeat template, a main and secondary class heartbeat detection unit 250 for detecting a main and secondary class heartbeat, and an supraventricular premature beat detection unit 260 for detecting a supraventricular premature beat;
the electrocardiograph signal reading unit 210 reads 5 minutes to-be-detected heartbeats and divides the 5 minutes to-be-detected heartbeats into 10 seconds data segments containing 2 seconds overlapping windows, all the 10 seconds data segments are respectively determined by the real-time premature beat detection unit 220 to be spliced into a preliminary detection result of the 5 minutes to-be-detected heartbeats, the 5 minutes to-be-detected data are respectively generated into a main template and a secondary template through the main template generation unit 230 and the secondary template generation unit 240 according to the preliminary detection result, then the main template and the secondary template are finally determined by the main and secondary heartbeat detection unit 250, and the ventricular premature beat detection unit 260 judges whether the heartbeat or the normal heartbeat on the basis of the normal heartbeat, wherein the specific steps are as follows: the electrocardiograph signal reading unit 210 extracts an electrocardiograph signal every 5 minutes, divides the 5 minute signal into a plurality of 10 second electrocardiograph data in a 2 second overlapping manner, and is used for detecting real-time ventricular premature beats and supraventricular premature beats in the real-time premature beat detection unit 220, the primary template generation unit 230 generates a primary template according to heartbeats with the largest variety in the detection result of the real-time premature beat detection unit 220, the secondary template generation unit 240 generates a secondary template according to heartbeats with the second largest variety in the detection result of the real-time premature beat detection unit 220, the primary and secondary cardiobeats detection unit 250 matches each heartbeat in the 5 minute electrocardiograph data with the primary and secondary templates generated by the primary template generation unit 230 and the secondary template generation unit 240, classifies the primary and secondary heartbeats according to the matching result, and the supraventricular premature beat detection unit 260 classifies the normal precordial beats detected by the primary and secondary cardiotomy detection unit 250 The RR interval of the postcardiotomy beat is related to the mean value of the RR interval of 5-minute electrocardiogram data, and the atrial fibrillation detection result is introduced, so that the detection of the supraventricular premature beat in the normal cardiotomy beat is realized.
Example 2
Periodic monitoring devices based on precordial beat signal among wearing formula electrocardiosignal, including wearing formula electrocardiosignal detection module 1 and long-time precordial beat detection module 2, long-time precordial beat detection module 2 can be implemented in wearing formula electrocardiosignal detection module 1's internal system, is the integration, for example: is arranged in the signal detection unit 130; meanwhile, the long premature beat detection module 2 may also be implemented in an external mobile platform, and is in a split type, for example, configured in a mobile terminal such as a mobile phone, as shown in fig. 2, in this embodiment, the long premature beat detection module 2 is in a split type and configured on the mobile phone.
Wearable electrocardiosignal detection module 1 accessible is wired or wireless mode and long-time premature beat heart beat detection module 2 is connected, simple structure, and the simple operation is nimble changeable, is applicable to different users' demand, wearable electrocardiosignal detection module 1 includes dry electrode unit 110, signal detection unit 130 and communication module unit 150, the number and the overall arrangement of dry electrode unit 110 are not unique, form the monitoring devices that different lead, and the wireless mode is selected for use to this embodiment and is connected, adopts 12 lead wearable electrocardiosignal detection module, and dry electrode unit pastes according to clinical 12 lead electrode and puts the position and arrange on human surface.
The 12-lead wearable electrocardiosignal detection module 1 acquires 12-lead electrocardiosignals through the dry electrode unit 110, converts analog signals into digital signals through the signal detection unit 130, sends the converted digital signals to the real-time premature beat detection module 2 through the communication module unit 150, and the electrocardiosignal reading unit 210, the real-time premature beat detection unit 220, the major-minor template generation unit 230, the minor template generation unit 240, the major-minor heartbeat detection unit 250 and the supraventricular premature beat detection unit 260 are sequentially connected, and the electrocardiosignal reading unit 210 extracts electrocardiosignals once every 5 minutes and divides the 5-minute signals into a plurality of 10-second electrocardiosignal data in a 2-second overlapping mode for detecting real-time ventricular premature beats and supraventricular premature beats in the real-time premature beat detection unit 220. As shown in fig. 3, the real-time premature beat detection unit 220 includes an electrocardiographic signal preprocessing step, an electrocardiographic signal feature extraction step, a quasi-premature beat detection step, a ventricular premature beat detection step, a supraventricular premature beat detection step, and a premature beat calibration step. The electrocardiosignal preprocessing link is used for preprocessing the extracted 10-second electrocardio data to obtain a preprocessed signal, the electrocardiosignal characteristic extraction link is used for extracting characteristic parameters such as an R wave position, an RR interval mean value, an R wave amplitude value, a QRS wave width and an R wave correlation coefficient of a heart beat to be detected from the preprocessed signal, and the quasi-premature beat detection link determines the quasi-premature beat by judging the RR interval; the ventricular premature beat detection link judges the R wave correlation coefficient, the QRS wave width and the R wave amplitude value to determine the ventricular premature beat on the basis of the quasi-premature beat; the supraventricular premature beat detection link obtains the supraventricular premature beat by removing the ventricular premature beat from the quasi-premature beat; and the premature beat checking link deletes the false detection beats in the ventricular premature beat and the supraventricular premature beat again through the R wave correlation coefficient, and searches whether the ventricular premature beat which is missed in detection exists in the normal beats. The main class template generating unit 230 generates a main class template according to the heart beats with the most variety in the detection result of the real-time premature beat detecting unit, the secondary template generating unit 240 generates a secondary template according to the heart beats with the second most variety in the detection result of the real-time premature beat detecting unit, the primary and secondary heartbeat detection unit matches each heartbeat in 5-minute electrocardiographic data with the primary and secondary templates generated by the primary template generation unit 230 and the secondary template generation unit 240, and classifying the primary and secondary heartbeats according to the matching result, wherein the supraventricular premature beat detection unit 260 introduces an atrial fibrillation detection result according to the relation between the RR intervals of the normal heartbeats and the RR intervals of the 5-minute electrocardiogram data before and after the normal heartbeats detected by the primary and secondary heartbeats detection unit 250, and realizes the detection of the supraventricular premature beats in the normal heartbeats.
The real-time premature beat heart beat detection module adopted by the embodiment is only aiming at simple characteristic parameters of the QRS waves such as RR interphase, R wave amplitude, QRS wave width, R wave correlation coefficient and the like, does not relate to too many morphological characteristics of the electrocardiographic wave, and has good processing effect and high detection accuracy aiming at dynamic electrocardiographic data with large baseline drift and much motion noise; meanwhile, an atrial fibrillation detection link is introduced, so that misjudgment of supraventricular premature beats is reduced, and the detection accuracy of the supraventricular premature beats is increased.
Example 3
The periodicity monitoring method based on the premature beat signal in the wearable electrocardiosignal comprises the following steps:
s1, reading electrocardiosignals: the electrocardiosignal is extracted once every 5 minutes, and the signal extracted within 5 minutes is divided into a plurality of 10-second electrocardio data in a mode of overlapping 2 seconds;
s2, detecting premature beat in real time: the steps are that the characteristic parameters of the heart beat to be detected are extracted, and parameter calculation and comparison are sequentially carried out on the characteristic parameters and the detection of the quasi-premature beat, the detection of the ventricular premature beat, the detection of the supraventricular premature beat and the verification of the premature beat, so that the monitoring and classification of the electrocardiosignals are realized, and the steps further comprise:
s21, preprocessing the electrocardiosignal;
s22, extracting electrocardiosignal characteristic parameters: extracting characteristic parameters of the heart beat to be detected from the electrocardiosignals preprocessed in the step S21, wherein the characteristic parameters comprise the R wave position of the heart beat to be detected in the 10-second electrocardiosignals, the RR intervals of all the heart beats to be detected, the RR interval mean value of the 10-second electrocardiosignals, the R wave amplitude of all the heart beats to be detected, the R wave amplitude mean value of the 10-second electrocardiosignals, the QRS wave width of all the heart beats to be detected, the QRS wave width mean value of the 10-second electrocardiosignals, the R wave correlation coefficients of all the heart beats to be detected and the R wave correlation number mean value of the 10-second electrocardiosignals;
the R wave position of the heart to be measured in the 10-second electrocardiosignal is used for positioning the position of the peak value of each R wave in the 10-second electrocardiosignal;
the RR interval of the heart beat to be measured is used for calculating the difference value between the R wave positions of the heart beat to be measured in the 10-second electrocardiosignal;
the average value of the RR intervals of the 10-second electrocardiosignals is used for calculating the average value of the RR intervals of the heart beats to be detected in the preprocessed 10-second electrocardiosignals;
the R wave amplitude of the heart beat to be measured is used for calculating the amplitude of the 10-second electrocardiosignal at the R wave position of the heart beat to be measured in the 10-second electrocardiosignal;
the average value of the R wave amplitude of the 10-second electrocardiosignal is used for calculating the average value of the amplitude of the 10-second electrocardiosignal at the R wave position of the heart to be measured in the 10-second electrocardiosignal;
the QRS wave width of the heart beat to be measured is used for calculating the QRS wave width of the 10-second electrocardiosignal at the R wave position of the heart beat to be measured in the 10-second electrocardiosignal;
the QRS wave width mean value of the 10-second electrocardiosignal is used for calculating the mean value of the QRS wave width of the 10-second electrocardiosignal at the R wave position of the heart beat to be measured in the 10-second electrocardiosignal;
the R wave correlation coefficient of the heart beat to be detected is used for acquiring the waveform of the heart beat to be detected from the 10-second electrocardiosignal and calculating the correlation coefficient among the waveforms of the heart beat to be detected;
the R wave correlation number average value of the 10-second electrocardiosignal is used for calculating the average value of the R wave correlation coefficient of the heart beat to be measured;
s23, judging the premature beat: judging whether the electrocardiosignal is a quasi-premature beat heartbeat according to the RR interval of the heartbeat to be detected, if so, continuing to step S24; if not, the signal is normal heartbeat, as shown in fig. 4, the step further includes:
s231, extracting the RR interphase mean value of the heart beat to be detected and the RR interphase mean value of the 10-second electrocardiosignal;
s232, starting from the second R wave position, judging whether the RR intervals before and after each R wave meet the condition that the former RR interval is smaller than the RR interval mean value of the 10-second electrocardiosignal, the latter RR interval is larger than the RR interval mean value of the 10-second electrocardiosignal, and the sum of the two RR intervals before and after is smaller than or equal to twice the RR interval mean value of the 10-second electrocardiosignal;
s233, if the result is yes, the current heartbeat to be tested is output as a quasi-premature beat, and the step S24 is continued; otherwise, if the result is negative, the current heartbeat to be detected is output as a normal heartbeat;
s24, judging ventricular premature beat: judging whether the quasi-premature beat screened in the step S23 is a ventricular premature beat or not according to the R wave correlation coefficient, the QRS wave width and the R wave amplitude of the beat to be measured, if not, screening out a non-ventricular premature beat, and continuing to the step S25; if yes, the signal is ventricular premature beat, and the determination method and the flow of the step are shown in fig. 5:
s241, extracting an RR interval of the quasi-premature beat heartbeat and an RR interval mean value of 10-second electrocardiogram data, an R wave correlation coefficient and an R wave correlation number mean value of 10-second electrocardiogram data, a QRS width and a QRS width mean value of 10-second electrocardiogram data, and an R wave amplitude mean value of 10-second electrocardiogram data;
s242, judging whether the R wave phase relation number of each quasi-premature beat heart beat is smaller than the mean value of the R wave phase relation numbers of the 10-second electrocardiogram data; if yes, go to S243 and S245 respectively; if not, the current quasi-premature beat output is a non-ventricular premature beat;
s243, judging whether the QRS width of the current quasi-premature beat heart beat is larger than the QRS width mean value of the 10-second electrocardiogram data; if so, go to S244; if not, the current quasi-premature beat output is a non-ventricular premature beat;
s244, judging whether the R wave amplitude of the current quasi-premature beat is abnormal to the R wave amplitude mean value of the 10-second electrocardiogram data; if so, outputting the current quasi-premature beat as a normal ventricular premature beat; if not, the current quasi-premature beat output is a non-ventricular premature beat;
whether the R wave amplitude of the current quasi-premature beat is abnormal to the R wave amplitude mean value of the 10-second electrocardiogram data or not refers to whether the R wave amplitude of the current quasi-premature beat is larger than or smaller than the R wave amplitude mean value of the 10-second electrocardiogram data or not, the larger or smaller selection depends on the relation between the mean value of the R wave amplitudes of all the quasi-premature beats and the R wave amplitude mean value of the 10-second electrocardiogram data, and if the mean value of the R wave amplitudes of all the quasi-premature beats is larger than the R wave amplitude mean value of the 10-second electrocardiogram data, the larger selection is carried out; otherwise, selecting less than;
s245, judging whether the sum of the RR intervals before and after the current quasi-premature beat heartbeat is 80-120% of the mean value of the RR intervals of the 10-second electrocardiosignals; if so, outputting the current quasi-premature beat as an insertion ventricular premature beat; if not, the current quasi-premature beat output is a non-ventricular premature beat;
s246, extracting R wave correlation coefficients of adjacent normal beats of the normal ventricular premature beat;
s247, judging whether the correlation coefficient is smaller than the R wave correlation number mean value of the 10-second electrocardiosignal and whether the correlation coefficient is between 80% and 120% of the R wave correlation number mean value of the 10-second electrocardiosignal; if yes, the current normal ventricular premature beat is adjacent and the output of the normal ventricular premature beat is a continuous ventricular premature beat; if not, the current normal ventricular premature beat is adjacently output as the normal beat;
s25, judging the supraventricular premature beat: judging whether the ventricular premature beat is the supraventricular premature beat or not according to the R wave correlation coefficient of the non-ventricular premature beat screened in the step S24 and the R wave correlation number average value of the 10-second electrocardiosignal, and if so, continuing to perform the step S26; if not, the signal is normal heartbeat;
judging whether the ventricular premature beat is the supraventricular premature beat or not according to the R wave correlation coefficient of the non-ventricular premature beat screened in the step S24 and the R wave correlation number average value of the 10-second electrocardiosignal, and if so, continuing to perform the step S26; if not, the signal is normal heartbeat; the supraventricular premature beat comprises two subtypes of a normal supraventricular premature beat and a continuous supraventricular premature beat, and the judging method and the flow are shown in figure 6:
s251, calculating the R wave correlation coefficient of the screened non-ventricular premature beat and the R wave correlation number mean value of the 10-second electrocardio data;
s252, judging whether the R wave phase relation number of each non-ventricular premature beat is larger than (0.8 times the mean value of the R wave phase relation numbers); if yes, the current non-ventricular premature beat output is the normal supraventricular premature beat, and the step S253 is continued; otherwise, if not, the current non-ventricular premature beat heart beat output is a normal heart beat;
s253, calculating the RR interval of the current normal supraventricular premature beat and the RR interval of the normal beat before the current normal supraventricular premature beat;
s254, judging whether the RR interval before the current normal supraventricular premature beat is larger than (0.8 times the average value of the R wave correlation number), wherein the RR interval is between 80% and 120% of the RR interval of the current normal supraventricular premature beat; if yes, outputting the normal heart beat before the current normal supraventricular premature heart beat and the current normal supraventricular premature heart beat as continuous supraventricular premature heart beats; otherwise, if not, the normal heartbeat output before the current normal supraventricular premature beat is the normal heartbeat;
s26, checking the premature beat and the heart beat: the premature beat check comprises a supraventricular premature beat self-check, a normal ventricular premature beat self-check and a ventricular premature beat mutual-check, and is used for judging whether a supraventricular premature beat has a false-check ventricular premature beat, whether a false-check noise signal exists in a normal ventricular premature beat and whether a missed-check ventricular premature beat exists in a normal beat, wherein the judging method and the flow are shown in figure 7: judging whether the number of the correlation coefficients between each supraventricular premature beat and all other supraventricular premature beats is less than 0.6 is 2/3 greater than the number of the other supraventricular premature beats, if so, the supraventricular premature beat is determined to be a normal ventricular premature beat; if the result is negative, determining the ventricular premature beat as the supraventricular premature beat;
the normal ventricular premature beat self-test judges whether a false-detected noise signal exists in the normal ventricular premature beat according to the correlation coefficient among the normal ventricular premature beats, and the judging method and the flow are shown as the following steps: judging whether the number of the correlation coefficients of each normal ventricular premature beat and other normal ventricular premature beats, which is less than 0.6, is greater than 2/3 of the number of the other normal ventricular premature beats, if so, the normal ventricular premature beat is determined to be a noise signal; if the result is negative, determining the ventricular premature beat as a normal ventricular premature beat;
the ventricular premature beat mutual detection judges whether the normal heart beat has a missed ventricular premature beat according to the correlation coefficient of the normal heart beat and the ventricular premature beat, and the judging method and the flow are shown as the following steps: judging whether the number of correlation coefficients of each normal heartbeat and the ventricular premature beats is larger than 2/3 of the number of the ventricular premature beats or not, if so, determining the normal heartbeat as the ventricular premature beats; if the result is negative, determining the heart beat as normal;
s3, generating a main class template: determining a main class template according to the premature beat detection result after the real-time premature beat detection in the step S2 and the correlation coefficient between heartbeats in the 5-minute data, as shown in fig. 10;
s31, extracting the detection results R1 of the ventricular premature beat and the supraventricular premature beat obtained after the real-time premature beat detection in the step S2;
s32, calculating the number of heartbeats of each type in R1, and generating a heartbeat P1 by taking the heartbeat with the largest number as a preparation main type template;
s33, calculating the correlation coefficient C1 among the heart beats in the preliminary main class template generation heart beat P1;
s34, sorting C1 in descending order, and selecting the heart beats ranked at the top 1/3 as a quasi-main template to generate heart beats P2;
s35, calculating the correlation coefficient C2 among the heart beats in the quasi-master template generation heart beat P2;
s36, sorting C2 in descending order, selecting the heart beats ranked at the top 1/3 as a main template to generate heart beats, and constructing a main template T1 according to the heart beats;
s37, calculating QRS width of the main template T1, and judging the type of the main template;
s4, generating a secondary template: determining a secondary template according to the premature beat detection result after the real-time premature beat detection in the step S2 and the correlation coefficient between beats in the 5-minute data, as shown in fig. 11;
s41, extracting the detection results R1 of the ventricular premature beat and the supraventricular premature beat obtained after the real-time premature beat detection in the step S2;
s42, calculating the number of heartbeats of each type in R1, and generating a heart beat P3 by using the heart beat with the second most number as a preparation secondary template;
s43, judging whether the number of the heart beats P3 generated by the preliminary class template is more than 10; if yes, go to step S44; if not, the output secondary template does not exist;
s44, calculating the correlation coefficient C3 among each heartbeat in the preliminary secondary template generation heartbeat P3;
s45, sorting C3 in descending order, and selecting the heart beats ranked at the top 1/3 as a quasi-class template to generate heart beats P4;
s46, calculating a correlation coefficient C4 among the heartbeats in the quasi-class template generation heartbeat P4;
s47, sorting C4 in descending order, selecting the heart beats ranked at the top 1/3 as a secondary template to generate heart beats, and constructing a primary template T2 according to the heart beats;
s48, determining the type of the minor class template T2 according to the type of the major class template T1;
the main class template and the secondary class template only have two conditions of normal heart beat, ventricular premature heart beat and normal heart beat;
s5, primary and secondary heartbeat detection: determining the type of each heartbeat according to the existence of the secondary template and the correlation coefficient between each heartbeat and the primary and secondary templates, as shown in fig. 12;
s51, extracting a main template T1 and a minor template T2;
s52, judging whether a minor template T2 exists; if so, continue to step S53; if not, go to step S57;
s53, calculating a correlation coefficient C5 between each heartbeat and the primary template and a correlation coefficient C6 between each heartbeat and the secondary template;
s54, judging whether the correlation coefficient C5 between the main type templates is more than 0.9; if so, continue to step S55; if not, go to step S56;
s55, judging whether the correlation coefficient C6 between the secondary templates is more than 0.9; if so, outputting noise in the current heartbeat type; if the result is negative, outputting the current heartbeat type as a main heartbeat type;
s56, judging whether the correlation coefficient C6 between the secondary templates is more than 0.9; if yes, outputting the current heartbeat type as a secondary heartbeat type; if the result is negative, outputting noise in the current heartbeat type;
s57, calculating a correlation coefficient C7 between each heartbeat and the main type template;
s58, judging whether the correlation coefficient C7 between the main type templates is more than 0.9; if so, outputting the current heartbeat type as a main heartbeat type; if the result is negative, outputting the current heartbeat type as a secondary heartbeat type;
s6, supraventricular premature beat detection: according to the characteristic parameters of the normal heartbeat and the atrial fibrillation detection result determined in the step S5, the supraventricular premature beat heartbeat is determined, as shown in fig. 13:
s61, extracting the RR interval mean values of the normal heart beat detected in the step S4, the RR interval mean values of the 5-minute electrocardiosignals and the RR interval mean values of the normal heart beat detected in the step S4;
s62, judging whether the RR interval before the normal heartbeat is less than the RR interval mean value of the electrocardiosignal of 5 minutes, whether the RR interval after the normal heartbeat is more than the RR interval mean value of the electrocardiosignal of 5 minutes, and whether the sum of the RR interval before the normal heartbeat and the RR interval after the normal heartbeat is less than or equal to twice the RR interval mean value of the electrocardiosignal of 5 minutes; if yes, go to step S63; if the result is negative, the current heart beat output is the normal heart beat;
s63, judging whether the 30-second electrocardiogram data of the current heart beat is an atrial fibrillation signal; if so, outputting the current heart beat as a normal heart beat; if not, the current heartbeat output is the supraventricular premature beat.
Therefore, after the judgment and monitoring are finished, the main template generating unit and the secondary template generating unit are adopted, the templates are generated by adopting a method of multiple matching screening, accurate template generation can be realized only through one characteristic parameter of the correlation coefficient on the basis of a real-time detection result, the accuracy of the templates is improved, and the algorithm is simple, rapid and accurate.
The foregoing shows and describes the general principles, essential features, and advantages of the invention. It will be understood by those skilled in the art that the present invention is not limited by the foregoing examples, which are provided to illustrate the principles of the invention, and that various changes and modifications may be made without departing from the spirit and scope of the invention, which is also intended to be covered by the appended claims. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (9)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811350454.6A CN109288515B (en) | 2018-11-14 | 2018-11-14 | Method and device for periodic monitoring of premature beats in wearable ECG signals |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811350454.6A CN109288515B (en) | 2018-11-14 | 2018-11-14 | Method and device for periodic monitoring of premature beats in wearable ECG signals |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109288515A CN109288515A (en) | 2019-02-01 |
| CN109288515B true CN109288515B (en) | 2021-06-22 |
Family
ID=65146663
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811350454.6A Active CN109288515B (en) | 2018-11-14 | 2018-11-14 | Method and device for periodic monitoring of premature beats in wearable ECG signals |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109288515B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111723622B (en) * | 2019-03-22 | 2024-04-26 | 安徽华米信息科技有限公司 | Heart beat classification method, heart beat classification device, wearable equipment and storage medium |
| CN112971790A (en) * | 2019-12-18 | 2021-06-18 | 华为技术有限公司 | Electrocardiosignal detection method, device, terminal and storage medium |
| EP3838133A1 (en) * | 2019-12-20 | 2021-06-23 | Kaptalia Monitoring | Device and process for ecg measurements |
| CN113768515A (en) * | 2021-09-17 | 2021-12-10 | 重庆邮电大学 | An ECG Signal Classification Method Based on Deep Convolutional Neural Networks |
| CN114052677B (en) * | 2021-11-19 | 2023-07-21 | 武汉联影智融医疗科技有限公司 | Pulse wave beat quality evaluation method, device, computer equipment and storage medium |
| CN114533081B (en) * | 2022-02-17 | 2023-08-25 | 东南大学 | Paroxysmal atrial fibrillation detection device fused with premature beat inhibition strategy |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101810476A (en) * | 2009-12-22 | 2010-08-25 | 李顶立 | Classification method of heart beat template of dynamic electrocardiogram |
| 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 |
| 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 |
| CN104905785A (en) * | 2015-06-25 | 2015-09-16 | 石家庄翰纬医疗设备有限公司 | Ventricular/ supraventricular premature heart beat detection method and system |
| CN107495960A (en) * | 2017-01-19 | 2017-12-22 | 山东医联万家健康科技有限公司 | A kind of heart real time signal monitoring processing method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7751876B2 (en) * | 2005-09-23 | 2010-07-06 | Hewlett-Packard Development Company, L.P. | Method and system for detecting premature ventricular contraction from a surface electrocardiogram |
-
2018
- 2018-11-14 CN CN201811350454.6A patent/CN109288515B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101810476A (en) * | 2009-12-22 | 2010-08-25 | 李顶立 | Classification method of heart beat template of dynamic electrocardiogram |
| 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 |
| 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 |
| CN104905785A (en) * | 2015-06-25 | 2015-09-16 | 石家庄翰纬医疗设备有限公司 | Ventricular/ supraventricular premature heart beat detection method and system |
| CN107495960A (en) * | 2017-01-19 | 2017-12-22 | 山东医联万家健康科技有限公司 | A kind of heart real time signal monitoring processing method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109288515A (en) | 2019-02-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109288515B (en) | Method and device for periodic monitoring of premature beats in wearable ECG signals | |
| Satija et al. | A review of signal processing techniques for electrocardiogram signal quality assessment | |
| US11517212B2 (en) | Electrocardiogram information dynamic monitoring method and dynamic monitoring system | |
| CN107358196B (en) | Heart beat type classification method and device and electrocardiograph | |
| CN106214145B (en) | Electrocardiogram classification method based on deep learning algorithm | |
| US11350868B2 (en) | Electrocardiogram information processing method and electrocardiogram workstation system | |
| US9724007B2 (en) | System and method for detecting a characteristic in an ECG waveform | |
| WO2019161609A1 (en) | Method for analyzing multi-parameter monitoring data and multi-parameter monitor | |
| WO2019161608A1 (en) | Multi-parameter monitoring data analysis method and multi-parameter monitoring system | |
| CN109411042B (en) | ECG information processing method and ECG workstation | |
| CN107714023A (en) | Static ecg analysis method and apparatus based on artificial intelligence self study | |
| CN108577830A (en) | A kind of user oriented sign information dynamic monitor method and dynamic monitor system | |
| CN107296600A (en) | A kind of quick 12 lead electrocardiosignal method for evaluating quality | |
| JP2024087811A (en) | Method and system for providing electrocardiogram measurement service using a portable electrocardiogram device | |
| Zhao et al. | PVC recognition for wearable ECGs using modified frequency slice wavelet transform and convolutional neural network | |
| CN109394206B (en) | Real-time monitoring method and device based on premature beat signal in wearable electrocardiosignal | |
| JP5933138B2 (en) | Apparatus and method for determining occurrence of QRS complex in ECG data | |
| Gu et al. | A novel diagnostic algorithm for heart disease in ECG monitoring system | |
| Venkatesh et al. | Can atrial lead system ameliorate the diagnosis of atrial arrhythmias? | |
| Mathe et al. | Advancements in Noise Reduction Techniques in ECG Signals: A Review | |
| Agrawal et al. | Morphological Event Based Signal Quality Assessment of Electrocardiogram | |
| Nasim | Adaptation of the Segmented Beat Modulation Method to support diagnosis of cardiovascular disorders using electrocardiographic tracings acquired by wearable sensors | |
| Islam et al. | Real-Time Acquisition and Classification of Electrocardiogram Signal | |
| CN114533081A (en) | Paroxysmal atrial fibrillation detection device fused with premature beat suppression strategy | |
| SEEMA et al. | ECG Based Cardiac Arrhythmia Detection using a Deep Neural Network |
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 |