CN109199361B - Method, storage medium and device for extracting R peak time in electrocardiosignal data - Google Patents
Method, storage medium and device for extracting R peak time in electrocardiosignal data Download PDFInfo
- Publication number
- CN109199361B CN109199361B CN201810788599.8A CN201810788599A CN109199361B CN 109199361 B CN109199361 B CN 109199361B CN 201810788599 A CN201810788599 A CN 201810788599A CN 109199361 B CN109199361 B CN 109199361B
- Authority
- CN
- China
- Prior art keywords
- sequence
- data
- interval
- waveform
- intervals
- 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.)
- Expired - Fee Related
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]
-
- 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
- A61B5/361—Detecting fibrillation
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Cardiology (AREA)
- Heart & Thoracic Surgery (AREA)
- Molecular Biology (AREA)
- Pathology (AREA)
- Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- Biophysics (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The invention provides a method for extracting R peak time in electrocardiosignal data, which comprises the steps of reading the first N data of the electrocardiosignal data, recording as data z (i), wherein i is 1,2 … N, the N data at least comprise one QRS wave, carrying out differential operation on z (i) to obtain a differential result dz (i), calculating the amplitude of z (i) to obtain an amplitude result Az (i); subdividing z (i) into K segments, TM(j) Is the difference maximum value, T, of the j-th stage difference result dz (i)m(j) Is the difference minimum value, T, of the j-th stage difference result dz (i)A(j) The maximum value of the amplitude of the j section amplitude result Az (i); updating a differential maximum
Differential minimum
And amplitude maximum
All dz (i) ≧ T are locatedMThe starting position of (d) is the sequence P1, all the alignment satisfying dz (i) ≦ TmIs the P2 sequence. The invention can accurately and quickly extract the R peak time in the electrocardiosignal data.
Description
Technical Field
The invention relates to the field of signal processing, in particular to a method, a storage medium and a device for extracting R peak time in electrocardiosignal data.
Background
Ventricular Fibrillation (VF, abbreviated as Ventricular Fibrillation) is a serious cardiovascular disease and is caused by coronary heart disease, myocardial infarction and other human causes or external causes such as surgery, drug poisoning and the like. In the event of ventricular fibrillation, the patient is often unconscious, pulseless, and blood pressure-free, in high risk, and at any time may be life threatening to the patient.
Timely detection and defibrillation are important means for treating patients recognized at present when the patients have ventricular fibrillation. If defibrillation can be performed within 1 minute of ventricular fibrillation occurring, the success rate of defibrillation can approach 100%, and if defibrillation occurs within 5 minutes of ventricular fibrillation, the success rate of defibrillation is reduced to 33%, and if effective defibrillation operation is still not performed within 10 minutes of ventricular fibrillation occurring, the success rate of defibrillation is almost zero. The judgment of the occurrence of ventricular fibrillation is a precondition for defibrillation treatment, and the ventricular fibrillation needs to be detected quickly and accurately in order to improve the success rate of defibrillation.
Cardiac electrical signals (ECG) are the basis for detecting ventricular fibrillation. If the doctor is relied on to manually check the electrocardiogram, judge whether ventricular fibrillation exists and carry out defibrillation, at least 5 minutes are needed, and therefore, the optimal time for ventricular fibrillation treatment is missed.
Therefore, it is urgently needed to develop a method for detecting ventricular fibrillation, automatically analyze a electrocardiosignal, automatically judge whether ventricular fibrillation occurs or not and improve the defibrillation success rate.
Disclosure of Invention
In view of this, the present invention provides a method, a storage medium, and a device for extracting R peak time in electrocardiographic signal data, so as to solve the problem of manually detecting ventricular fibrillation.
The invention provides a method for extracting R peak time in electrocardiosignal data, which comprises the following steps:
reading the first N data of the electrocardiosignal data, and recording as z (i), wherein i is 1 and 2 … N, the N data at least comprise one QRS wave, performing difference operation on z (i) to obtain dz (i), and calculating the amplitude of z (i) to be Az (i);
subdividing z (i) into K segments, where TM(j) Is the difference maximum value, T, of the j-th section dz (i)m(j) Is the difference minimum value, T, of the j section dz (i)A(j) Is the maximum amplitude value of the j section Az (i); updating
All dz (i) ≧ T are locatedMThe starting position of (d) is the sequence P1, all the alignment satisfying dz (i) ≦ TmIs a P2 sequence,
if the P1 sequence and the P2 sequence are both non-empty, judging whether a matching P2 exists in each P1 in the P1 sequence item by item, wherein the matching P2 belongs to the P2 sequence and is separated from the corresponding P1 by less than one QRS wave interval; if so, determining whether there is Az (i) ≧ T between the P1 and the matching P2AIf so, extracting Az (i) as the R peak time corresponding to the maximum value between the P1 and the matching P2.
The present invention also provides a non-transitory computer readable storage medium storing instructions that, when executed by a processor, cause the processor to perform the steps in the ventricular fibrillation detection method of the present application.
The invention also provides a ventricular fibrillation detection device, which is characterized by comprising a processor and the non-transitory computer-readable storage medium.
The invention provides a method for extracting R peak time in electrocardiosignal data, and when the detection method is used for collecting the electrocardiosignal data in real time, ventricular fibrillation can be judged quickly and accurately in real time, a basis is provided for timely defibrillation, and the defibrillation success rate is improved.
Drawings
FIG. 1 is a graph of a typical cardiac signal;
FIG. 2 is a flow chart of the ventricular fibrillation detection method of the present invention;
FIG. 3 is one embodiment of step 103 of FIG. 2;
FIG. 4 is one embodiment of step 1031 in FIG. 3;
FIG. 5 is the first embodiment of FIG. 2;
FIG. 6 is a second embodiment of FIG. 2;
FIG. 7 is a third embodiment of FIG. 2;
FIG. 8 is a fourth embodiment of FIG. 2;
FIG. 9 is a fifth embodiment of FIG. 2;
FIG. 10 is a sixth embodiment of FIG. 2;
FIG. 11 is one embodiment of the extraction of RR intervals of FIGS. 5-10;
FIG. 12 is one embodiment of step 1011 of FIG. 11;
fig. 13 is a structural view of the ventricular fibrillation detection apparatus of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in detail with reference to the accompanying drawings and specific embodiments.
The ventricular fibrillation detection method analyzes electrocardiosignal data, and as shown in fig. 1, an electrocardiosignal graph is a typical electrocardiosignal graph, and the electrocardiosignal graph is composed of three main characteristic waves, namely a P wave, a QRS wave and a T wave. Where the P-wave is the first occurring waveform, reflecting atrial depolarization, the P-wave duration typically does not exceed 0.11 seconds. The QRS wave reflects ventricular depolarization and is composed of a Q wave, an R wave and an S wave, and the duration time is generally 0.06-0.10 seconds. Wherein, Q wave and S wave direction are downward, R wave direction is upward, and R wave is the most obvious waveform of peak value in an electrocardio cycle. The T wave reflects ventricular repolarization, following the QRS wave. The PR interval is the time from the beginning of the P wave to the beginning of the QRS wave, and the QT interval is the time from the beginning of the QRS wave to the end of the T wave. The RR interval is the time interval between two adjacent R peaks, or the time required for the onset of one ventricular depolarization to the onset of the next ventricular depolarization.
The RR interval is also called heart rate interval, and the conversion formula of the heart rate and the RR interval is:
heart rate 60/RR interval
The unit of the heart rate is 'times/minute', the unit of the RR interval is 'seconds', and the judgment of whether an R peak exists in the electrocardiosignal data and the positioning of the position of the R peak are the key points for obtaining the RR interval and the heart rate data.
As shown in fig. 2, the method for detecting ventricular fibrillation in the present invention includes:
step 101: sequentially extracting characteristic information of the electrocardiosignal data, and executing a step 103 when the continuously extracted characteristic information in a first preset time meets a first judgment condition, wherein the first judgment condition is a distinguishing characteristic of the characteristic information when the ventricular fibrillation exists and the characteristic information when the ventricular fibrillation does not exist;
in step 101, the electrocardiographic signal data includes non-real-time changing data and real-time changing data, where the non-real-time changing data may be historical data, or data automatically stored or cached after the data acquisition device acquires data for a fixed time (e.g., 6 seconds); the real-time changing data may be real-time updated data or buffered data recorded while collecting.
For the data that changes in non-real time, the implementation manner of extracting the feature information and judging the feature information in step 101 may be to extract all information data in the electrocardiographic signal data and then judge, or judge every time one or more feature information is extracted.
And dynamically extracting one or more characteristic information of the electrocardiosignal data which changes in real time according to the acquisition time of the data and then judging.
It should be noted that, in the present invention, the dynamic extraction of the feature information, the dynamic storage of the feature information, or the dynamic judgment of the feature information are all performed in sequence, that is, the dynamic extraction is performed according to the sequence of data time, the data collected first is analyzed first, and then the data collected is analyzed later.
The characteristic information is information obtained by waveform detection identification and parameter extraction on the electrocardiographic signal data, and the information is obviously different between the absence of ventricular fibrillation and the occurrence of ventricular fibrillation, such as one or more parameters in fig. 1.
Preferably, the RR interval data and the waveform characteristics of the electrocardiograph signals are selected as the characteristic information in step 101, and are easy to identify or extract, and have obvious difference before and after ventricular fibrillation occurs.
The waveform characteristics are used for indicating the oscillation state of the electrocardiosignal data. The waveform characteristics include: waveform information of the electrocardiosignal data, or inflection point information of the electrocardiosignal data, or local peak information of the electrocardiosignal data, wherein the waveform information can be the position of a waveform or data time corresponding to the waveform; the inflection point information may be a position of an inflection point or a data time corresponding to the inflection point; the local peak information may be the location of the local peak or the time of the data corresponding to the local peak. In unit time, the more the number of waveforms, the more the number of inflection points or the more the number of local peaks, the more the electrocardiographic signal waveform oscillates.
Assuming that z (n) is the electrocardiographic signal data of step 101, the method for calculating the inflection point includes:
step A-1: calculating a difference dz (n) for z (n), wherein dz (n) is z (n) z (n-1).
Step A-2: comparing whether the sign of the differential value dz (n) at the current time and the sign of the differential value dz (n-1) at the previous time are the same, if the signs of the two are different, namely dz (n) dz (n-1) <0, recording dz (n) or dz (n-1) as an inflection point.
The first preset duration is set to both efficiency and reliability of detection, and the longer the first preset duration is set, the higher the reliability of the result is, but at the same time, the detection time is increased, and the detection efficiency is reduced, for example, the first preset duration may be set to 20 seconds.
Preferably, the first preset time period is 10 seconds to 20 seconds. Because the success rate of defibrillation can approach 100% if defibrillation can be performed in real time within 1 minute of ventricular fibrillation, considering the time of subsequent calculation and operation, when the first preset time is 10 seconds to 20 seconds, defibrillation can be performed within 1 minute of ventricular fibrillation.
When ventricular fibrillation occurs, the RR interphase and waveform characteristics of the electrocardiosignal are different from normal data or characteristics, the first judgment condition of the RR interphase can be set according to the RR interphase data characteristics during ventricular fibrillation, the first judgment condition of the waveform characteristics can also be set according to the waveform characteristics during ventricular fibrillation, and the method is not limited by the application. For example, the RR interval may be determined by referring to a numerical range or a converted heart rate range, and the waveform characteristics may be determined by selecting the number of waveforms or the number of inflection points in a unit time.
On the other hand, if the RR interval or the waveform feature does not accord with the first judgment condition within the first preset time length, the extracted RR interval and the extracted waveform feature are continuously detected.
Step 103: calculating the complexity of the electrocardiosignal data within a first preset time length, judging whether the complexity is greater than a first preset value, and if so, judging that ventricular fibrillation is detected; otherwise, judging that no ventricular fibrillation is detected.
The first preset value is related to a first preset time, and different first preset times correspond to different first preset values.
Because the electrocardiosignal data is easy to mix noise or interference signals, the reliability of the analysis result of the step 101 is affected, that is, when the characteristic information meets the first judgment condition, ventricular fibrillation may not occur.
Therefore, after the characteristic information is determined, step 103 is added, wherein the complexity reflects the rate of new patterns appearing in a time sequence along with the increase of the length of the time sequence, and the degree of the sequence approaching to randomness is shown. The complexity of the electrocardiosignal data is calculated to distinguish whether the electrocardiosignal data belongs to ventricular fibrillation data characteristics or random data characteristics, and the complexity larger than a first preset value indicates that the electrocardiosignal data belongs to the ventricular fibrillation data characteristics.
The ventricular fibrillation determination method based on the steps 101 and 103 not only considers the characteristic information of the electrocardiosignals, but also considers the complexity characteristic of the electrocardio data, so that the ventricular fibrillation can be determined quickly and accurately by the method, and the defibrillation success rate is improved.
In step 103, the complexity of calculating the electrocardiosignal data within the first preset time period may be calculated by using the prior art, and also using the method shown in fig. 3:
step 1031: converting the electrocardiosignal data within the first preset time length into a binary sequence;
step 1032: and calculating the complexity of the binary sequence by adopting Lempel-Ziv.
Further, as shown in fig. 4, step 1031 may further include:
step 1031-1: let the electrocardio signal data in the first preset time period be z (n), n be 1,2 … L, and calculate the average value of z (n) as z (n)m(ii) a Statistics of z (n)>zmNumber of E, z (n)<zmThe number of (2) is F;
step 1031-2: if E + F<α L, then H ═ zm(ii) a If E + F is not less than alpha L and E is<F, then H ═ β E + zm(ii) a If E + F is not less than alpha L and E is not less than F, then H is not less than beta F + zm，0<ɑ、β<1；
Step 1031-3: and comparing z (n) with H item by item, if z (n) > H, setting z (n) to 1, otherwise setting z (n) to 0.
The values of α and β may be set with reference to a complexity characteristic of the history data, for example, α is set to 0.5 and β is set to 0.2.
When the characteristic information is the RR interval and the waveform feature, the specific implementation manner of "sequentially extracting and determining the characteristic information" in step 101 of fig. 2 may be to determine the RR interval and then determine the waveform feature, or determine the waveform feature and then determine the RR interval, or determine the RR interval and the waveform feature at the same time. Examples of these three implementations are given below, respectively.
Example one and example two
One embodiment is shown in fig. 5.
Step 201: sequentially extracting RR intervals in the electrocardiosignal data, and storing the newly extracted RR intervals into a newly recorded RR interval sequence;
step 202: when the new recorded RR interval sequence is not empty, sequentially taking out RR intervals in the new recorded RR interval sequence, and when the probability that the RR intervals continuously taken out within a first preset time length are in an abnormal range is greater than a second preset value, executing a step 203;
step 203: and identifying the waveform characteristics of the electrocardiosignal data within the first preset duration, executing the step 103 if the waveform characteristics accord with a preset oscillation state, otherwise, judging that ventricular fibrillation is not detected, and returning to the step 202.
Step 103: calculating the complexity of the electrocardiosignal data within a first preset time length, judging whether the complexity is greater than a first preset value, and if so, judging that ventricular fibrillation is detected; otherwise, judging that no ventricular fibrillation is detected, and returning to the step 202.
The above steps 201 and 202 may be run in parallel, where both steps are directed to "newly recorded RR interval sequence", the step 201 dynamically extracts RR intervals and stores the RR intervals in "newly recorded RR interval sequence", and the step 202 is directed to "newly recorded RR interval sequence", and when the "newly recorded RR interval sequence" is not empty, the data in the sequence is taken out and the judgment is performed, so that the "newly recorded RR interval sequence" is a dynamically changing sequence, and is dynamically stored and taken out.
In the presence of ventricular fibrillation, the data for RR intervals may deviate from normal values, i.e. appear as outliers, e.g. extracted RR intervals such as RRaThe value of (A) is an abnormal value (in an abnormal range), which may be the case in RRaBefore, the data values taken out are all normal values, then from RRaInitially, data of the RR intervals subsequently taken out are continuously monitored for a first preset time period, and if most of the RR interval data in the time period are in an abnormal range (that is, the probability that the RR intervals continuously taken out in the first preset time period are in the abnormal range is greater than a second preset value), step 203 is executed. The second preset value mainly considers the ratio of abnormal RR interval data in the first preset duration to all RR interval data in the duration so as to ensure that the abnormal RR interval is not a sporadic event. For example, the second preset value may be set to 90% or more.
The RR interval data includes time information or position information of the data, so that the start-stop time or start-stop position of the electrocardiographic signal data corresponding to the RR interval data of the first preset duration can be determined according to the time information or position information of the start-stop RR interval within the first preset duration, thereby determining the data range of the waveform feature identification of step 203.
On the other hand, if the RR interval is in the abnormal range probability less than or equal to the second preset value within the first preset time, continuing to take out data from the new recorded RR interval sequence and executing judgment.
In step 203, the preset oscillation state may be an inflection point or a local peak or a number of waveforms within a first preset time period greater than or equal to a selected value.
Example two is shown in figure 6.
Step 201-2: sequentially extracting RR intervals in the electrocardiosignal data, and storing the newly extracted RR intervals into an RR interval sequence, wherein the newly extracted RR intervals are the RR intervals which are not analyzed in the RR interval sequence;
step 202-2: when the RR intervals which are not analyzed in the RR interval sequence are not empty, reading the RR intervals which are not analyzed in sequence, converting the RR intervals which are not analyzed into the analyzed RR intervals after being read, and executing a step 203-2 when the probability that the RR intervals which are continuously read within a first preset time length are in an abnormal range is greater than a second preset value;
step 203-2: and identifying the waveform characteristics of the electrocardiosignal data within the first preset duration, executing the step 103 if the waveform characteristics accord with the preset oscillation state, otherwise, judging that ventricular fibrillation is not detected, and returning to the step 202-2.
Step 103: calculating the complexity of the electrocardiosignal data within a first preset time length, judging whether the complexity is greater than a first preset value, and if so, detecting ventricular fibrillation; otherwise, no ventricular fibrillation is detected and the process returns to step 202-2.
An unanalyzed RR interval in the sequence of RR intervals refers to an RR interval that has not been detected by step 202-2. In a specific implementation, the identifier may be set to distinguish between the analyzed RR interval data, for example, the identifier of the analyzed RR interval data is "1", and the flag symbol of the analyzed RR interval is "0"; or the latest data time or data position corresponding to the analyzed RR interval data is recorded, and the RR interval data after the data time or the data position belongs to the analyzed RR interval data.
Since the RR interval sequence is constantly updated by step 201-2, the RR interval sequence is a real-time updated sequence, such as RR1、RR2、…RRa、RRa+1…. In step 202-2, assume that a currently read RR interval, e.g., RRaThe value of (A) is an abnormal value (in an abnormal range), which may be the case in RRaPreviously, if the data values read were all normal values, then the RR is followedaInitially, the data of the RR intervals read subsequently is continuously monitored for a first preset time period, and if most of the RR interval data is in an abnormal range in the time period, step 203 is executed.
The implementation manner of step 101 in fig. 5 and fig. 6 is to determine RR intervals first and then determine waveform characteristics.
Example three and example four
Example three is shown in fig. 7.
Step 301: sequentially extracting waveform characteristics in the electrocardiosignal data, and storing the newly extracted waveform characteristics into a newly recorded waveform characteristic sequence;
step 302: when the newly recorded waveform characteristic sequence is not empty, sequentially taking out the waveform characteristics in the newly recorded waveform characteristic sequence, and when the waveform characteristics continuously taken out within a first preset time length accord with a preset oscillation state, executing a step 303;
step 303: and (3) extracting an RR interval of the electrocardiosignal data in the first preset time, if the RR interval is in an abnormal range and the probability is greater than a second preset value, executing the step 103, otherwise, judging that ventricular fibrillation is not detected, and returning to the step 302.
Step 103: calculating the complexity of the electrocardiosignal data within a first preset time length, judging whether the complexity is greater than a first preset value, and if so, judging that ventricular fibrillation is detected; otherwise, judging that no ventricular fibrillation is detected, and returning to the step 302.
Example four is shown in fig. 8.
Step 301-2: sequentially extracting waveform characteristics in the electrocardiosignal data, and storing the newly extracted waveform characteristics into a waveform characteristic sequence, wherein the newly extracted waveform characteristics are waveform characteristics which are not analyzed in the waveform characteristic sequence;
step 302-2: when the unanalyzed waveform features in the waveform feature sequence are not empty, sequentially reading the unanalyzed waveform features, converting the unanalyzed waveform features into analyzed waveform features after the unanalyzed waveform features are read, and executing the step 303-2 when the waveform features continuously read within a first preset time length accord with a preset oscillation state;
step 303-2: and (3) extracting an RR interval of the electrocardiosignal data in the first preset time, if the RR interval is in an abnormal range and the probability is greater than a second preset value, executing the step 103, otherwise, judging that ventricular fibrillation is not detected, and returning to the step 302-2.
Step 103: calculating the complexity of the electrocardiosignal data within a first preset time length, judging whether the complexity is greater than a first preset value, and if so, judging that ventricular fibrillation is detected; otherwise, judging that no ventricular fibrillation is detected, and returning to the step 302-2.
The implementation manner of step 101 in fig. 7 and 8 is to determine the waveform characteristics first and then determine the RR intervals.
Example five and example six
Example five is shown in figure 9.
Step 401: sequentially extracting RR intervals in the electrocardiosignal data and waveform characteristics of each RR interval, and storing the newly extracted RR intervals into a newly recorded RR interval sequence; storing the newly extracted waveform characteristics into a newly recorded waveform characteristic sequence;
step 402: when the newly recorded RR interval sequence or the newly recorded waveform feature sequence is not empty, sequentially taking out RR intervals in the newly recorded RR interval sequence and waveform features corresponding to the RR intervals, and when the probability that the continuously taken RR intervals are in an abnormal range within a first preset time is greater than a second preset value and the corresponding waveform features also accord with a preset oscillation state, executing step 103;
step 103: calculating the complexity of the electrocardiosignal data within a first preset time length, judging whether the complexity is greater than a first preset value, and if so, detecting ventricular fibrillation; otherwise, no ventricular fibrillation is detected and the process returns to step 402.
Example six is shown in fig. 10.
Step 401-2: sequentially extracting RR intervals in the electrocardiosignal data and waveform characteristics of each RR interval, and storing the newly extracted RR intervals to an RR interval sequence; storing the newly extracted waveform characteristics to a waveform characteristic sequence; the newly extracted waveform features are waveform features which are not analyzed in the waveform feature sequence; the newly extracted RR interval is an unanalyzed RR interval in the RR interval sequence;
step 402-2: when the RR intervals of the RR interval sequence are not analyzed or the waveform features of the RR interval sequence are not empty, sequentially reading the RR intervals of the RR interval sequence which are not analyzed and the waveform features corresponding to the RR intervals, converting the RR intervals of the RR interval sequence which are not analyzed into the analyzed RR intervals after being read, converting the waveform features of the RR intervals which are not analyzed into the analyzed waveform features after being read, and executing step 103 when the RR intervals which are continuously read within a first preset time period are in an abnormal range probability which is greater than a second preset value and the corresponding waveform features also accord with preset oscillation states;
step 103: calculating the complexity of the electrocardiosignal data within a first preset time length, judging whether the complexity is greater than a first preset value, and if so, detecting ventricular fibrillation; otherwise, no ventricular fibrillation is detected and the process returns to step 402-2.
In addition, the invention also improves the method for extracting the RR interval, and as shown in FIG. 11, the method comprises the following steps:
step 1011: extracting R peak moments of the electrocardiosignal data one by one according to time sequence, and calculating a time interval delta t between the newly extracted R peak moment and the last recorded R peak moment;
step 1012: judging whether deltat is greater than the refractory period duration and less than an RR interval threshold, wherein the RR interval threshold is a preset multiple (for example, 1.5 times or 2 times) of the maximum value of the currently recorded RR interval, if so, executing the step 1013, otherwise, returning to the step 1011;
if the current RR interval is empty, the RR interval threshold assumes a default value.
Step 1013: and recording the newly extracted R peak time, and calculating the time interval between the newly recorded R peak time and the last recorded R peak time as the newly extracted RR interval.
Further, as shown in fig. 12, the method includes: step 1011 further includes:
step 1011-1: reading the first N data of the unanalyzed electrocardiosignal data, and recording as z (i), wherein i is 1,2 … N, N data at least comprises one QRS wave, performing difference operation on z (i) to obtain dz (i), and calculating the amplitude of z (i) to be Az (i);
step 1011-2: judging whether the electrocardiosignal data is initially detected, if so, executing a step 1011-3, otherwise, executing a step 1011-4;
for example, by judging TM、TmAnd TAWhether the detection is the initial detection or not is judged by the default value or not, or an initial detection label is set.
Step 1011-3: subdividing z (i) into K segments, where TM(j) Is the difference maximum value, T, of the j-th section dz (i)m(j) Is the difference minimum value, T, of the j section dz (i)A(j) Is the maximum amplitude value of the j section Az (i); updating
Step 1011-4 is performed;
step 1011-4: all dz (i) ≧ T are locatedMThe starting position of (d) is the sequence P1, all the alignment satisfying dz (i) ≦ TmIs a P2 sequence, if both P1 and P2 are not null, then step 1011-5 is executed, otherwise, step 1011-1 is returned to;
step 1011-5: judging whether a matching P2 exists in each P1 in the P1 sequence item by item, wherein the matching P2 belongs to the P2 sequence and the interval between the matching P2 and the corresponding P1 is less than one QRS wave interval; if so, continue to determine if there is Az (i) ≧ T between P1 and the matching P2AIf not, returning to step 1011-1, if yes, extracting Az (i) and taking the data time corresponding to the maximum value between P1 and matching P2 as the R peak time, after the P1 sequence detection is finished, if the detection is the initial detection, returning to step 1011-1, and if the detection is not the initial detection, returning to step 1011-1 after step 1011-3 is executed.
On the other hand, the electrocardiosignal data mentioned in fig. 2 to 11 are obtained by filtering the acquired initial electrocardiosignal data, and the filtering includes low-pass filtering and high-pass filtering.
Let the initial cardiac signal data be x (j), the output of the low-pass filtering be y (j), and then y (n) and x (n) satisfy:
y(j)＝2y(j-1)–y(j-2)+x(j)–2x(j-6)+x(j-12)
let the output of the high-pass filtering be z (j), then z (j) and y (j) satisfy:
z(j)＝-z(j-1)+32y(j-16)–y(j)+y(j-32)
the inventors also provide a non-transitory computer readable storage medium storing instructions, characterized in that the instructions, when executed by a processor, cause the processor to perform the steps in the inventive method of detecting ventricular fibrillation as described in any one of the above.
The inventor also provides a ventricular fibrillation detection device, which is characterized by comprising a processor and the non-transitory computer-readable storage medium.
As shown in fig. 13, the ventricular fibrillation detection apparatus of the present invention includes:
the characteristic information extraction and judgment module: extracting feature information of the electrocardiosignal data, and executing a complexity extraction and judgment module when the feature information continuously extracted within a first preset time duration meets a first judgment condition, wherein the first judgment condition is a distinguishing feature of the feature information when ventricular fibrillation exists and when ventricular fibrillation does not occur;
a complexity extraction and judgment module: calculating the complexity of the electrocardiosignal data within a first preset time length, judging whether the complexity is greater than a first preset value, and if so, judging that ventricular fibrillation is detected; otherwise, judging that no ventricular fibrillation is detected.
Preferably, the characteristic information includes RR intervals and waveform characteristics.
Optionally, the characteristic information extracting and judging module includes:
a new recorded RR interval sequence updating module: extracting RR interphase in the electrocardiosignal data, and storing the newly extracted RR interphase into a newly recorded RR interphase sequence;
RR interval judging module 1: when the new recorded RR interval sequence is not empty, sequentially taking out RR intervals in the new recorded RR interval sequence, and when the probability that the RR intervals continuously taken out within a first preset time length are in an abnormal range is greater than a second preset value, executing a waveform characteristic judgment module 1;
the waveform characteristic judging module 1: and identifying the waveform characteristics of the electrocardiosignal data within the first preset duration, executing the complexity extraction and judgment module if the waveform characteristics accord with the preset oscillation state, otherwise, judging that ventricular fibrillation is not detected, and returning to the RR interphase sequence judgment module 1.
Optionally, the characteristic information extracting and judging module includes:
the RR interval sequence updating module: extracting RR intervals in the electrocardiosignal data, storing the newly extracted RR intervals into an RR interval sequence, wherein the newly extracted RR intervals are the RR intervals which are not analyzed in the RR interval sequence;
RR interval judging module 2: when the RR intervals which are not analyzed in the RR interval sequence are not empty, reading the RR intervals which are not analyzed in sequence, converting the RR intervals which are not analyzed into the analyzed RR intervals after being read, and executing a waveform characteristic judgment module 2 when the probability that the RR intervals which are continuously read within a first preset time length are in an abnormal range is greater than a second preset value;
the waveform characteristic judging module 2: and identifying the waveform characteristics of the electrocardiosignal data within the first preset duration, executing the complexity extraction and judgment module if the waveform characteristics accord with the preset oscillation state, otherwise, judging that ventricular fibrillation is not detected, and returning to the RR interphase sequence judgment module 2.
Optionally, the characteristic information extracting and judging module includes:
the newly recorded waveform characteristic sequence updating module: extracting waveform characteristics in the electrocardiosignal data, and storing the newly extracted waveform characteristics into a newly recorded waveform characteristic sequence;
the waveform characteristic judging module 3: when the newly recorded waveform characteristic sequence is not empty, sequentially taking out the waveform characteristics in the newly recorded waveform characteristic sequence, and when the waveform characteristics continuously taken out within a first preset time length accord with a preset oscillation state, executing an RR interval judgment module 3;
RR interval judging module 3: and (3) extracting an RR interval of the electrocardiosignal data in the first preset time, if the RR interval is in an abnormal range and the probability is greater than a second preset value, executing a complexity extraction and judgment module, and if not, judging that ventricular fibrillation is not detected, and returning to the waveform characteristic judgment module 3.
Optionally, the characteristic information extracting and judging module includes:
a waveform characteristic sequence updating module: extracting waveform characteristics in the electrocardiosignal data, storing the newly extracted waveform characteristics into a waveform characteristic sequence, wherein the newly extracted waveform characteristics are waveform characteristics which are not analyzed in the waveform characteristic sequence;
the waveform characteristic judging module 4: when the unanalyzed waveform features in the waveform feature sequence are not empty, the unanalyzed waveform features are read in sequence, the unanalyzed waveform features are converted into the analyzed waveform features after being read, and when the waveform features continuously read within a first preset time length accord with a preset oscillation state, an RR interval judgment module 4 is executed;
RR interval judging module 4: and (3) extracting an RR interval of the electrocardiosignal data in the first preset time, if the RR interval is in an abnormal range and the probability is greater than a second preset value, executing a complexity extraction and judgment module, and if not, judging that ventricular fibrillation is not detected, and returning to the waveform characteristic judgment module 4.
Optionally, the characteristic information extracting and judging module includes:
a new recording sequence updating module: extracting RR intervals in the electrocardiosignal data and waveform characteristics of each RR interval, and storing the newly extracted RR intervals into a newly recorded RR interval sequence; storing the newly extracted waveform characteristics into a newly recorded waveform characteristic sequence;
the characteristic information judging module 1: and when the RR interval sequence or the waveform feature sequence is not empty, sequentially taking out the RR interval in the RR interval sequence and the waveform feature corresponding to the RR interval, and executing the complexity extraction and judgment module when the probability that the RR interval continuously taken out within a first preset time is in an abnormal range is greater than a second preset value and the corresponding waveform feature also accords with a preset oscillation state.
Optionally, the characteristic information extracting and judging module includes:
a sequence update module: extracting RR intervals in the electrocardiosignal data and waveform characteristics of each RR interval, and storing the newly extracted RR intervals to an RR interval sequence; storing the newly extracted waveform characteristics to a waveform characteristic sequence; the newly extracted waveform features are waveform features which are not analyzed in the waveform feature sequence; the newly extracted RR interval is an unanalyzed RR interval in the RR interval sequence;
the characteristic information judging module 2: when the RR intervals which are not analyzed in the RR interval sequence or the waveform features which are not analyzed in the waveform feature sequence are not empty, sequentially reading the RR intervals which are not analyzed in the RR interval sequence and the waveform features corresponding to the RR intervals, converting the RR intervals which are not analyzed into the analyzed RR intervals after being read, converting the waveform features which are not analyzed into the analyzed waveform features after being read, and executing the complexity extraction and judgment module when the RR intervals which are continuously read within a first preset time length are in abnormal range probability which is greater than a second preset value and the corresponding waveform features also accord with preset oscillation states.
Preferably, the first preset time is 10-20 seconds, the first preset value is related to the first preset time, and the second preset value is greater than or equal to 90%.
Preferably, the calculating the complexity of the electrocardiosignal data within the first preset time duration comprises:
a binary conversion module: converting the electrocardiosignal data within the first preset time length into a binary sequence;
a complexity extraction module: and calculating the complexity of the binary sequence by adopting Lempel-Ziv.
Further, the binary conversion module includes:
a statistic module: let the electrocardio signal data in the first preset time period be z (n), n be 1,2 … L, and calculate the average value of z (n) as z (n)m(ii) a Statistics of z (n)>zmNumber of E, z (n)<zmThe number of (2) is F;
a reference value calculation module: if E + F<α L, then H ═ zm(ii) a If E + F is not less than alpha L and E is<F, then H ═ β E + zm(ii) a If E + F is not less than alpha L and E is not less than F, then H is not less than beta F + zm，0<ɑ、β<1；
A binary conversion module: and comparing z (n) with H item by item, if z (n) > H, setting z (n) to 1, otherwise setting z (n) to 0.
Further, α ═ 0.5 and β ═ 0.2.
Optionally, the waveform characteristics include: the waveform information of the electrocardiosignal data, or the inflection point information of the electrocardiosignal data, or the local peak information of the electrocardiosignal data.
Optionally, extracting RR intervals in the cardiac electrical signal data comprises:
an R peak extraction module: extracting R peak moments of the electrocardiosignal data one by one according to time sequence, and calculating a time interval delta t between the newly extracted R peak moment and the last recorded R peak moment;
an R peak judgment module: judging whether delta t is greater than the refractory period duration and smaller than an RR interval threshold, wherein the RR interval threshold is a preset multiple of the maximum value of the currently recorded RR interval, if so, executing an R peak recording module, and otherwise, returning to the R peak extracting module;
r peak recording module: and recording the newly extracted R peak time, and calculating the time interval between the newly recorded R peak time and the last recorded R peak time as the newly extracted RR interval.
Optionally, the R peak extraction module further comprises:
electrocardiosignal data reading module: reading the first N data of the unanalyzed electrocardiosignal data, and recording as z (i), wherein i is 1,2 … N, N data at least comprises one QRS wave, performing difference operation on z (i) to obtain dz (i), and calculating the amplitude of z (i) to be Az (i);
a threshold value judging module: judging whether the electrocardiosignal data is initially detected, if so, executing a threshold updating module and then executing a P1 and P2 positioning module, otherwise, directly executing a P1 and P2 positioning module;
a threshold updating module: subdividing z (i) into K segments, where TM(j) Is the difference maximum value, T, of the j-th section dz (i)m(j) Is the difference minimum value, T, of the j section dz (i)A(j) Is the maximum amplitude value of the j section Az (i); updating
P1, P2 positioning module: all dz (i) ≧ T are locatedMThe starting position of (d) is the P1 sequence, all dz (i) ≦ TmThe initial position of the R peak locating module is a P2 sequence, if the P1 sequence and the P2 sequence are both non-null, the R peak locating module is executed, otherwise, the R peak locating module returns to the electrocardiosignal data reading module;
an R peak localization module: judging whether a matching P2 exists in each P1 in the P1 sequence item by item, wherein the matching P2 belongs to the P2 sequence and the interval between the matching P2 and the corresponding P1 is less than one QRS wave interval; if so, continue to determine if there is Az (i) ≧ T between P1 and the matching P2AIf the data point (b) is the maximum value between the P1 and the matching P2, extracting Az (i), wherein the data time corresponding to the maximum value between the P1 and the matching P2 is the R peak time, after the P1 sequence detection is finished, if the data point is the initial detection, directly returning to the electrocardiosignal data reading module, and if the data point is not the initial detection, executing a threshold valueAnd the module is updated and then returns to the electrocardiosignal data reading module.
Optionally, the acquired initial electrocardiograph signal data is filtered to obtain electrocardiograph signal data, and the filtering includes low-pass filtering and high-pass filtering.
Let the initial cardiac signal data be x (j), the output of the low-pass filtering be y (j), and the following conditions are satisfied between y (n) and x (n): y (j) -2 y (j-1) -y (j-2) + x (j) -2 x (j-6) + x (j-12); let the output of the high-pass filtering be z (j), and the following are satisfied between z (j) and y (j): z (j) ═ z (j-1) +32y (j-16) -y (j) + y (j-32).
It should be noted that the embodiments of the ventricular fibrillation detection apparatus according to the present invention have the same principle as the embodiments of the ventricular fibrillation detection method, and the relevant points can be referred to each other.
The above description is only exemplary of the present invention and should not be taken as limiting the scope of the present invention, and any modifications, equivalents, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (14)
1. A method for extracting R peak time in electrocardiosignal data is characterized by comprising the following steps:
reading the first N data of the electrocardiosignal data, and recording the data as data z (i), wherein i is 1, 2.. N, the N data at least comprises one QRS wave, performing difference operation on z (i) to obtain a difference result dz (i), and calculating the amplitude of z (i) to obtain an amplitude result Az (i);
subdividing z (i) into K segments, TM(j) Is the difference maximum value, T, of the j-th stage difference result dz (i)m(j) Is the difference minimum value, T, of the j-th stage difference result dz (i)A(j) The maximum value of the amplitude of the j section amplitude result Az (i); updating a differential maximum
Differential minimum
And amplitude maximum
All dz (i) ≧ T are locatedMThe starting position of (d) is the sequence P1, all the alignment satisfying dz (i) ≦ TmIs a P2 sequence,
if the P1 sequence and the P2 sequence are both non-empty, judging whether a matching P2 exists in each P1 in the P1 sequence item by item, wherein the matching P2 belongs to the P2 sequence and is separated from the corresponding P1 by less than one QRS wave interval; if the data point exists, judging whether a data point of Az (i) ≧ TA exists between the P1 and the matching P2, and if the data point of Az (i) is extracted, wherein the data time corresponding to the maximum value between the P1 and the matching P2 is the R peak time.
2. The method of claim 1, wherein prior to subdividing z (i) into K segments, further comprising,
judging whether the electrocardiosignal data is initially detected or not, if so, executing the step of subdividing z (i) into K segments,
otherwise, performing the positioning all satisfying dz (i) ≧ TMThe starting position of (d) is the sequence P1, all the alignment satisfying dz (i) ≦ TmIs the step of the P2 sequence.
3. The method of claim 1, further comprising,
and if the P1 sequence and the P2 sequence are both non-null, executing the step of reading the first N data of the electrocardiosignal data.
4. The method of claim 1, further comprising,
after the P1 sequence is detected, if the detection is initial detection, executing the step of reading the first N data of the electrocardiosignal data, and if the detection is not initial detection, executing the step of positioning all the data (dz) (i) is not less than TMThe starting position of (c) is the P1 sequence, all of which are aligned with dz: (i)≤TmIs the step of the P2 sequence.
5. The method of claim 1, further comprising,
step 1011: extracting R peak moments of the electrocardiosignal data one by one according to time sequence, and calculating a time interval delta t between the newly extracted R peak moment and the last recorded R peak moment;
step 1012: judging whether the delta t is greater than the refractory period duration and smaller than an RR interval threshold, wherein the RR interval threshold is a preset multiple of the maximum value of the currently recorded RR interval, if so, executing the step 1013, and otherwise, returning to the step 1011;
step 1013: and recording the newly extracted R peak time, and calculating the time interval between the newly recorded R peak time and the last recorded R peak time as the newly extracted RR interval.
6. The method of claim 5, further comprising sequentially extracting waveform features of the cardiac signal data, wherein the waveform features are indicative of an oscillation state of the cardiac signal data.
7. The method of claim 6, further comprising,
step 101: based on the extracted characteristic information of the electrocardiosignal data, when the characteristic information continuously extracted within a first preset time duration meets a first judgment condition, executing a step 103, wherein the first judgment condition is a distinguishing characteristic of the characteristic information when ventricular fibrillation exists and when ventricular fibrillation does not occur;
step 103: and calculating the complexity of the electrocardiosignal data in the first preset time length, wherein the characteristic information comprises RR intervals and waveform characteristics.
8. The method of claim 7,
the step 101 comprises:
step 201: sequentially extracting RR intervals in the electrocardiosignal data, and storing the newly extracted RR intervals into a newly recorded RR interval sequence;
step 202: when the new recorded RR interval sequence is not empty, sequentially taking out RR intervals in the new recorded RR interval sequence, and when the probability that the RR intervals continuously taken out within a first preset time length are in an abnormal range is greater than a second preset value, executing a step 203;
step 203: identifying the waveform characteristics of the electrocardiosignal data within the first preset time, executing the step 103 if the waveform characteristics accord with a preset oscillation state, otherwise, returning to the step 202;
or
The step 101 comprises:
step 201-2: sequentially extracting RR intervals in electrocardiosignal data, and storing newly extracted RR intervals into an RR interval sequence, wherein the newly extracted RR intervals are the RR intervals which are not analyzed in the RR interval sequence;
step 202-2: when the RR intervals which are not analyzed in the RR interval sequence are not empty, reading the RR intervals which are not analyzed in sequence, converting the RR intervals which are not analyzed into the analyzed RR intervals after being read, and executing a step 203-2 when the probability that the RR intervals which are continuously read within a first preset time length are in an abnormal range is greater than a second preset value;
step 203-2: identifying the waveform characteristics of the electrocardiosignal data within a first preset time length, executing the step 103 if the waveform characteristics accord with a preset oscillation state, otherwise, returning to the step 202-2;
or
The step 101 comprises:
step 301: sequentially extracting waveform characteristics in the electrocardiosignal data, and storing the newly extracted waveform characteristics into a newly recorded waveform characteristic sequence;
step 302: when the newly recorded waveform characteristic sequence is not empty, sequentially taking out the waveform characteristics in the newly recorded waveform characteristic sequence, and when the waveform characteristics continuously taken out within a first preset time length accord with a preset oscillation state, executing a step 303;
step 303: extracting an RR interval of the electrocardiosignal data in the first preset time, if the RR interval is in an abnormal range, the probability is greater than a second preset value, executing step 103, otherwise, returning to step 302;
or
The step 101 comprises:
step 301-2: sequentially extracting waveform characteristics in electrocardiosignal data, and storing the newly extracted waveform characteristics into a waveform characteristic sequence, wherein the newly extracted waveform characteristics are waveform characteristics which are not analyzed in the waveform characteristic sequence;
step 302-2: when the unanalyzed waveform features in the waveform feature sequence are not empty, sequentially reading the unanalyzed waveform features, converting the unanalyzed waveform features into analyzed waveform features after the unanalyzed waveform features are read, and executing the step 303-2 when the waveform features continuously read within a first preset time length accord with a preset oscillation state;
step 303-2: extracting an RR interval of the electrocardiosignal data in the first preset time, if the RR interval is in an abnormal range, the probability is greater than a second preset value, executing the step 103, otherwise, returning to the step 302-2;
or
The step 101 comprises:
step 401: sequentially extracting RR intervals in the electrocardiosignal data and waveform characteristics of each RR interval, and storing the newly extracted RR intervals into a newly recorded RR interval sequence; storing the newly extracted waveform characteristics into a newly recorded waveform characteristic sequence;
step 402: when the new recorded RR interval sequence or the new recorded waveform feature sequence is not empty, sequentially taking out RR intervals in the new recorded RR interval sequence and waveform features corresponding to the RR intervals, and when the probability that the RR intervals continuously taken out within a first preset time length are in an abnormal range is greater than a second preset value and the corresponding waveform features also accord with a preset oscillation state, executing step 103;
or
The method step 101 comprises:
step 401-2: sequentially extracting RR intervals in the electrocardiosignal data and waveform characteristics of each RR interval, and storing the newly extracted RR intervals into an RR interval sequence; storing the newly extracted waveform characteristics into a waveform characteristic sequence; the newly extracted waveform features are waveform features which are not analyzed in the waveform feature sequence; the newly extracted RR interval is an unanalyzed RR interval in the RR interval sequence;
step 402-2: when the RR intervals of the RR interval sequence are not analyzed or the waveform features of the RR intervals of the waveform feature sequence are not empty, sequentially reading the RR intervals of the RR interval sequence that are not analyzed and the waveform features corresponding to the RR intervals, converting the RR intervals of the RR interval sequence that are not analyzed into the analyzed RR intervals after the RR intervals of the RR interval sequence that are not analyzed are read, converting the waveform features of the RR intervals that are not analyzed into the analyzed waveform features after the RR intervals of the RR interval sequence that are not analyzed are read, and executing step 103 when the RR intervals that are continuously read within a first preset time period are in an abnormal range probability that is greater than a second preset value and the corresponding waveform features also conform to a preset oscillation state.
9. The method of claim 8, wherein the first predetermined time period is 10 seconds or less and 20 seconds or less, and the second predetermined value is 90% or more.
10. The method of claim 7, wherein the calculating the complexity of the first predetermined duration of the cardiac signal data comprises:
step 1031: converting the electrocardiosignal data within the first preset time length into a binary sequence;
step 1032: and calculating the complexity of the binary sequence by adopting Lempel-Ziv.
11. The method according to claim 10, wherein the step 1031 comprises:
step 1031-1: let the electrocardiosignal data in the first preset time period be z (n), n be 1,2 … L, and calculate the average value of z (n) as zm(ii) a Statistics of z (n)>zmNumber of E, z (n)<zmThe number of (2) is F;
step 1031-2: if E + F<α L, then used for the assigned variable H ═ zm(ii) a If E + F is not less than alpha L and E is<F, then H ═ β E + zm(ii) a If E + F is not less than alpha L and E is not less than F, thenH＝βF+zmWherein, 0<ɑ、β<1；
Step 1031-3: and comparing z (n) with H item by item, if z (n) > H, setting z (n) to 1, otherwise setting z (n) to 0.
12. The method according to claim 1, wherein the cardiac signal data is obtained by low-pass filtering and high-pass filtering the collected initial cardiac signal data;
let the initial cardiac signal data be x (i), the output of the low-pass filtering be y (i), and then the following conditions are satisfied between y (i) and x (i): y (i) -2 y (i-1) -y (i-2) + x (i) -2 x (i-6) + x (i-12);
let the output of the high-pass filtering be z (i), then the following is satisfied between z (i) and y (i): z (i) ═ z (i-1) +32y (i-16) -y (i) + y (i-32).
13. A non-transitory computer readable storage medium storing instructions, wherein the instructions, when executed by a processor, cause the processor to perform the steps in the method for extracting R peak time in electrocardiographic signal data according to any one of claims 1 to 12.
14. An apparatus for extracting an R-peak time in electrocardiographic signal data, comprising a processor and the non-transitory computer-readable storage medium according to claim 13.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810788599.8A CN109199361B (en) | 2018-07-18 | 2018-07-18 | Method, storage medium and device for extracting R peak time in electrocardiosignal data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810788599.8A CN109199361B (en) | 2018-07-18 | 2018-07-18 | Method, storage medium and device for extracting R peak time in electrocardiosignal data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109199361A CN109199361A (en) | 2019-01-15 |
| CN109199361B true CN109199361B (en) | 2022-03-04 |
Family
ID=64990031
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810788599.8A Expired - Fee Related CN109199361B (en) | 2018-07-18 | 2018-07-18 | Method, storage medium and device for extracting R peak time in electrocardiosignal data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109199361B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113100779B (en) * | 2020-01-10 | 2024-07-05 | 深圳市理邦精密仪器股份有限公司 | Ventricular fibrillation detection method, device and monitoring equipment |
| CN112914589B (en) * | 2021-03-02 | 2023-04-18 | 钦州市第二人民医院 | Multi-sleep-guidance monitoring wireless net cap device and monitoring method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5545186A (en) * | 1995-03-30 | 1996-08-13 | Medtronic, Inc. | Prioritized rule based method and apparatus for diagnosis and treatment of arrhythmias |
| US6490478B1 (en) * | 2000-09-25 | 2002-12-03 | Cardiac Science Inc. | System and method for complexity analysis-based cardiac tachyarrhythmia detection |
| CN1989897A (en) * | 2005-12-29 | 2007-07-04 | 深圳迈瑞生物医疗电子股份有限公司 | Ventricular fibrillation combined detecting method based on complexity |
-
2018
- 2018-07-18 CN CN201810788599.8A patent/CN109199361B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5545186A (en) * | 1995-03-30 | 1996-08-13 | Medtronic, Inc. | Prioritized rule based method and apparatus for diagnosis and treatment of arrhythmias |
| US6490478B1 (en) * | 2000-09-25 | 2002-12-03 | Cardiac Science Inc. | System and method for complexity analysis-based cardiac tachyarrhythmia detection |
| CN1989897A (en) * | 2005-12-29 | 2007-07-04 | 深圳迈瑞生物医疗电子股份有限公司 | Ventricular fibrillation combined detecting method based on complexity |
Non-Patent Citations (1)
| Title |
|---|
| Detecting Ventricular Tachycardia and Fibrillation by Complexity Measure;Xu-Sheng Zhang et.al.;《IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING》;19990531;第46卷(第5期);548-555 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109199361A (en) | 2019-01-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103110417B (en) | Automatic electrocardiogram recognition system | |
| EP3801238B1 (en) | Detecting abnormalities in ecg signals | |
| KR102451795B1 (en) | ECG signal detection method | |
| CN111772628B (en) | An automatic detection system for ECG signal atrial fibrillation based on deep learning | |
| Yeh et al. | Analyzing ECG for cardiac arrhythmia using cluster analysis | |
| CN109700450B (en) | Heart rate detection method and electronic equipment | |
| CN109770860B (en) | Electrocardiosignal processing device and electrocardio equipment | |
| CN109171712A (en) | Auricular fibrillation recognition methods, device, equipment and computer readable storage medium | |
| CN107456227B (en) | Full-lead electrocardiogram clustering template system and method | |
| CN109758145B (en) | An automatic sleep staging method based on the causal relationship of EEG signals | |
| US7881779B2 (en) | Method and apparatus for synthetically detecting ventricular fibrillation | |
| WO2007043903A1 (en) | Method, device and system for lead-limited electrocardiography (ecg) signal analysis | |
| CN103654770A (en) | Method and device for real-time detection of moving electrocardiosignal QRS (magnetic resonance angiography) waves | |
| CN111358459A (en) | Arrhythmia identification method, device, equipment and storage medium | |
| CN110693488A (en) | ECG signal processing system, method, electronic device and readable storage medium | |
| CN111832537B (en) | Abnormal electrocardiosignal identification method and abnormal electrocardiosignal identification device | |
| CN111956207B (en) | Electrocardiogram recording labeling method, device, equipment and storage medium | |
| CN113951893A (en) | A multi-lead ECG feature point extraction method combining deep learning and electrophysiological knowledge | |
| CN109199361B (en) | Method, storage medium and device for extracting R peak time in electrocardiosignal data | |
| CN105877739A (en) | Clinical examination method of electrocardio intelligent analyzing system | |
| US7764987B2 (en) | Signal processing apparatus for physiological signals | |
| CN116173412A (en) | Defibrillation equipment control method and device based on rhythm recognition and defibrillation equipment | |
| CN110960207A (en) | Tree model-based atrial fibrillation detection method, device, equipment and storage medium | |
| CN108836312B (en) | Clutter rejection method and system based on artificial intelligence | |
| CN116725486A (en) | Dynamic electrocardiosignal atrial fibrillation detection method and device based on double phases |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20220304 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |