CN116304777B - Self-adaptive electrocardiosignal denoising method and system based on reference signal during rest - Google Patents
Self-adaptive electrocardiosignal denoising method and system based on reference signal during rest Download PDFInfo
- Publication number
- CN116304777B CN116304777B CN202310385919.6A CN202310385919A CN116304777B CN 116304777 B CN116304777 B CN 116304777B CN 202310385919 A CN202310385919 A CN 202310385919A CN 116304777 B CN116304777 B CN 116304777B
- Authority
- CN
- China
- Prior art keywords
- signal
- filtered
- reference signal
- denoising
- coefficient
- 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
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000001914 filtration Methods 0.000 claims abstract description 41
- 230000001133 acceleration Effects 0.000 claims abstract description 17
- 230000003068 static effect Effects 0.000 claims abstract description 11
- 230000003044 adaptive effect Effects 0.000 claims description 9
- 230000011218 segmentation Effects 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 5
- 230000000747 cardiac effect Effects 0.000 claims description 3
- 230000035945 sensitivity Effects 0.000 claims description 3
- 230000000694 effects Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 3
- 238000012952 Resampling Methods 0.000 description 2
- 239000000654 additive Substances 0.000 description 2
- 230000000996 additive effect Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000000354 decomposition reaction Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000002245 particle Substances 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 208000024172 Cardiovascular disease Diseases 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 206010003119 arrhythmia Diseases 0.000 description 1
- 230000006793 arrhythmia Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 230000001747 exhibiting effect Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003183 myoelectrical effect Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000000284 resting effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/318—Heart-related electrical modalities, e.g. electrocardiography [ECG]
- A61B5/346—Analysis of electrocardiograms
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7203—Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal
- A61B5/7207—Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal of noise induced by motion artifacts
- A61B5/721—Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal of noise induced by motion artifacts using a separate sensor to detect motion or using motion information derived from signals other than the physiological signal to be measured
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
- G06F2218/04—Denoising
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
Abstract
The invention discloses a self-adaptive electrocardiosignal denoising method and system based on a reference signal in a static state, and the method comprises the following steps: acquiring an acceleration signal based on the accelerometer; judging a motion state based on the acceleration signal; distinguishing a reference signal from a signal to be filtered based on the motion state; in a multi-threshold SWT denoising stage, dynamically setting a threshold value by calculating the relation between the reference signal and the signal to be filtered; in the similarity filtering stage, dynamically setting coefficients by calculating the relation between the reference signal and the signal to be filtered; and denoising the signal to be filtered based on the threshold value and the coefficient. The invention takes the electrocardiosignals of the person when the person is stationary as the reference signal, and can accurately judge the noise intensity in real time by comparing the electrocardiosignals of the person when the person is stationary with the current electrocardiosignals, thereby dynamically setting the threshold value and the coefficient, and ensuring that the algorithm has stronger capability of processing the variation noise.
Description
Technical Field
The invention belongs to the technical field of signal processing, and particularly relates to a self-adaptive electrocardiosignal denoising method and system based on a reference signal in a static state.
Background
Accurate detection of Electrocardiograph (ECG) is of great interest for preventing cardiovascular disease, assisting in recovery of a group of slow patients, etc. However, in the ECG detection process of daily life, particularly in application scenarios such as exercise health monitoring by using wearable equipment, ECG signals face pollution of noise such as baseline drift, exercise artifacts, electromyographic signals, and the like, which seriously affects accurate judgment and prevention of ECG related diseases. Traditional denoising filtering algorithms, such as Kalman Filter (KF) and wiener filtering, are all based on the assumption of a linear system to process signals, and perform poorly in the face of nonlinear and non-stationary signals such as electrocardiograph signals. Although some improvements such as Extended KF (EKF) have been developed to enhance the ability of the algorithm to process nonlinear signals, the dynamic adaptation of the algorithm is limited because the algorithm requires a prior modeling of the signal. Other filtering algorithms such as Non-local Mean (NLM) learn intrinsic features common to electrocardiographic waveforms by comparing multiple segments of electrocardiographic signals, but tend to ignore some unique features of each segment.
The most widely adopted methods at present are various time-frequency analysis algorithms such as empirical mode decomposition (Empirical Mode Decomposition, EMD), stable wavelet transformation (Stationary Wavelet Transform, SWT) and the like, which remove various noises by decomposing signals onto different spatial scales and setting different thresholds for the signals of the different spatial scales, but the performance of the algorithm is extremely dependent on the selection of the thresholds and has weak adaptability in the face of the changed noises.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides the self-adaptive electrocardiosignal denoising method and system based on the reference signal at rest, which takes the electrocardiosignal of a person at rest as the reference signal, and can accurately judge the noise intensity in real time by comparing the electrocardiosignal at rest with the current electrocardiosignal, thereby dynamically setting the threshold value and coefficient, and enabling the algorithm to have stronger capability of processing the change noise.
In order to achieve the above object, the present invention provides the following solutions:
the self-adaptive electrocardiosignal denoising method based on the reference signal in the static state comprises the following steps of:
acquiring an acceleration signal based on the accelerometer;
judging a motion state based on the acceleration signal;
distinguishing a reference signal and a signal to be filtered based on the motion state, wherein the reference signal and the signal to be filtered are electrocardiosignals;
in a multi-threshold SWT denoising stage, dynamically setting a threshold value by calculating the relation between the reference signal and the signal to be filtered;
in the similarity filtering stage, dynamically setting coefficients by calculating the relation between the reference signal and the signal to be filtered;
and denoising the signal to be filtered based on the threshold value and the coefficient.
Preferably, the signal to be filtered is preprocessed before the multi-threshold SWT denoising stage, and the preprocessing method includes: low pass filtering, median filtering and signal segmentation;
wherein, the low pass filtering is: filtering out signals above 80 Hz;
the median filtering is as follows: calculating the median value of the signals to be filtered in each section of window by adopting a sliding window, and subtracting the median value from the signals to be filtered to remove baseline drift;
the signal segmentation is as follows: the signal is segmented by cardiac cycle and resampled to the same length.
Preferably, the multi-threshold SWT denoising method includes:
dividing an input signal into a P, T, U wave part with low frequency and a QR and RS wave band part with high frequency;
reserving a high-frequency QR wave band part and a high-frequency RS wave band part;
for the P, T, U wave part with low frequency, 7-level detail coefficient D is obtained through 7-level SWT 1 -D 7 And the final approximation coefficient A, the corresponding wavelet base selects "sym6";
direct removal of detail coefficient D 1 -D 2 The method comprises the steps of carrying out a first treatment on the surface of the Preserving detail coefficient D 6 -D 7 And a final approximation coefficient a; by a predetermined threshold value for detail coefficient D 3 -D 5 Distinguishing between them.
Preferably, the detail factor D is determined by a predetermined threshold 3 -D 5 The distinguishing method comprises the following steps:
dividing the signals into reference signals and signals to be denoised according to the motion state, wherein the set of the reference signals is denoted as Q;
calculating the average standard deviation of the reference signal coefficientsStandard deviation of signal coefficient to be denoised +.>And corresponding ratio r i ；
Wherein the average standard deviationThe expression of (2) is:
standard deviation of the signal coefficient to be denoisedThe expression of (2) is:
the corresponding ratio r i The expression of (2) is:
preferably, for a signal of length N, the threshold corresponds to the expression th i And a thresholding function D i，de The following are provided:
wherein, alpha is used for adjusting the sensitivity degree to noise, and 0.2 is taken.
Preferably, the method for similarity filtering includes:
classifying signals by adopting a K-means algorithm, and carrying out non-local mean filtering on the current signal Z (n) to be filtered in each class S:
calculating the weight w (U, Z) of each signal U in the class S relative to the current signal Z (n) to be filtered; all signals are weighted and summed to obtain a filtered signal
Calculating the energy of the signal to be filtered and the average energy E of the reference signal F A ratio R of (2);
based on the current signal Z (n) to be filtered and the filtered signalAnd the ratio R, the final filtered signal is obtained.
Preferably, the filtered signalThe expression of (2) is:
preferably, based on the preprocessed signal F (n), the ratio R is expressed as:
preferably, the expression of the final filtered signal is:
the invention also provides a self-adaptive electrocardiosignal denoising system based on the reference signal in the static state, which comprises the following steps: the system comprises a signal acquisition module, a judgment module, a distinguishing module, a threshold setting module, a coefficient setting module and a denoising module;
the signal acquisition module is used for acquiring an acceleration signal based on the accelerometer;
the judging module is used for judging the motion state based on the acceleration signal;
the distinguishing module is used for distinguishing a reference signal and a signal to be filtered based on the motion state;
the threshold setting module is used for dynamically setting a threshold by calculating the relation between the reference signal and the signal to be filtered in a multi-threshold SWT denoising stage;
the coefficient setting module is used for dynamically setting coefficients by calculating the relation between the reference signal and the signal to be filtered in the similarity filtering stage;
the denoising module is used for denoising the signal to be filtered based on the threshold value and the coefficient.
Compared with the prior art, the invention has the beneficial effects that:
the invention provides a self-adaptive electrocardiosignal denoising method and system based on a reference signal in static state, which take the electrocardiosignal of a person in static state as the reference signal, and can accurately judge the noise intensity in real time by comparing the electrocardiosignal in static state with the current electrocardiosignal, thereby dynamically setting a threshold value and a coefficient and having stronger capability of processing variable noise. The invention can well remove noise and simultaneously maintain the characteristics of signal waveforms, and has optimal performance when facing different input signal to noise ratios, and can better adapt to the changed noise intensity.
Drawings
In order to more clearly illustrate the technical solutions of the present invention, the drawings that are needed in the embodiments are briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flowchart of an adaptive electrocardiosignal denoising method based on a stationary reference signal in an embodiment of the invention;
FIG. 2 is a flow chart of preprocessing in an embodiment of the present invention;
FIG. 3 is a flowchart of multi-threshold SWT denoising in an embodiment of the present invention;
FIG. 4 is a flow chart of similarity filtering in an embodiment of the invention;
FIG. 5 is a graph showing the comparison of denoising effects according to an embodiment of the present invention;
fig. 6 is a diagram showing the denoising effect on a real noisy signal in an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In order that the above-recited objects, features and advantages of the present invention will become more readily apparent, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description.
Example 1
As shown in fig. 1, the invention provides a self-adaptive electrocardiosignal denoising method based on a reference signal in a static state, which comprises the following steps:
acquiring an acceleration signal based on the accelerometer;
judging a motion state based on the acceleration signal;
distinguishing a reference signal and a signal to be filtered based on a motion state, wherein the reference signal and the signal to be filtered are electrocardiosignals;
in a multi-threshold SWT denoising stage, dynamically setting a threshold value by calculating the relation between a reference signal and a signal to be filtered;
in the similarity filtering stage, dynamically setting coefficients by calculating the relation between the reference signal and the signal to be filtered;
based on the threshold and the coefficient, denoising of the signal to be filtered is completed.
In this embodiment, the acceleration signal is used to determine which part of the signal is the resting electrocardiographic signal, and can be used as a reference signal. The stationary judgment standard is that the acceleration signal of more than half of the time in one electrocardiographic period is less than 0.1g.
In this embodiment, as shown in fig. 2, the signal to be filtered is preprocessed before the multi-threshold SWT denoising stage, where the preprocessing includes a low-pass filtering, a median filtering, and a signal segmentation. The low-pass filtering filters out signals above 80Hz, the median filtering adopts a sliding window to calculate the median value of the signals to be filtered in each section of window, and the median value is subtracted by the signals to be filtered, so that the baseline drift is removed. Signal segmentation segments the signal by cardiac cycle (R peak-R peak) and resamples to the same length. If the sampling signal frequency is fs (360 Hz in this embodiment), the sliding window size is set to 0.6fs and the resampling length is set to 0.7fs.
In this embodiment, as shown in fig. 3, the method for denoising the multi-threshold SWT includes:
dividing an input signal into a P, T, U wave part with low frequency and a QR and RS wave band part with high frequency;
the high-frequency part signal is not easily affected by noise, and is not processed at first;
for the P, T, U wave part with low frequency, 7-level detail coefficient D is obtained through 7-level SWT 1 -D 7 And the final approximation coefficient A, the corresponding wavelet base selects "sym6";
for detail coefficient D 1 -D 2 This part mainly represents high-frequency noise, and can be directly and completely removed; for detail coefficient D 6 -D 7 And the final approximation coefficient a represents mainly the signal portion, which is preserved here; and detail coefficient D 3 -D 5 Representing noise and signal aliasing, a thresholding function D is required by subsequent calculations i，de Distinguishing between them.
In the present embodiment, the threshold processing function D obtained by the subsequent calculation i，de For detail coefficient D 3 -D 5 The distinguishing method comprises the following steps:
dividing the signals into reference signals and signals to be denoised according to the motion state, wherein the set of the reference signals is denoted as Q;
calculating the average standard deviation of the reference signal coefficientsStandard deviation of signal coefficient to be denoised +.>And corresponding ratio r i ；
Wherein the average standard deviationThe expression of (2) is:
standard deviation of the signal coefficient to be denoisedThe expression of (2) is:
the corresponding ratio r i The expression of (2) is:
r i the lower the ratio, the stronger the influence of noise, and the larger the corresponding threshold.
In the present embodiment, for a signal of length N (length 256 in the present embodiment), the threshold corresponds to expression th i The following are provided:
where α is used to regulate the sensitivity to noise, typically 0.2. The coefficients of each stage after denoising can be obtained by the following threshold processing function.
The next denoising process is then performed after combination by Inverse stationary wavelet transform (Inverse SWT, ISWT) numbers.
In this embodiment, as shown in fig. 4, the method for similarity filtering includes:
firstly, signals are divided into reference signals and signals to be denoised, for the signals to be denoised, because normal electrocardio waveforms and abnormal electrocardio waveforms exist in electrocardio signals, and the abnormal electrocardio waveforms account for a small number, in order to avoid overlarge weight caused by overlarge factors when a large number of normal electrocardio waveforms calculate weights, filtering the abnormal electrocardio signals is influenced, the signals are classified by adopting a K-means algorithm (the class number is 12 in the embodiment), and non-local mean filtering (NLM filtering) is carried out on the current signals Z (n) to be filtered in each class S:
calculating the weight w (U, Z) of each signal U in the class S relative to the current signal Z (n) to be filtered; the weight reflects the degree of similarity between signals, the more similar the signals are, the greater the weight is;
all signals are weighted and summed to obtain a filtered signal
In this embodiment, the filtered signalThe expression of (2) is:
but after NLM filtering, the common characteristics of signals are more preserved, and the unique characteristics of some signals are lost. When the signal itself is not greatly affected by noise, the strength of the filtering can be reduced. Therefore, based on the preprocessed signal F (n), the ratio R of the energy of the signal to be filtered to the average energy EF of the reference signal is calculated; since noise mainly affects the P, T, U waves, i.e. approximately the interval 0.1fs-0.6fs in the signal, only the energy of this part is calculated.
In this embodiment, the expression of the ratio R is:
when R is smaller, which means that the noise is more, the more the result of NLM filtering is adopted, and the more R is, which means that the noise contained in the signal is not more, the noise removal part of the multi-threshold SWT is enough to remove the noise, so that the more the output signal part of multi-threshold SWT denoising is adopted, and the final filtered signal is as follows:
after resampling the signal to the original length, the final output signal is obtained.
In this embodiment, the performance of the algorithm of the present invention is first intuitively reflected by exhibiting the effect of denoising the electrocardiograph signal, the data set being an MIT-BIH arrhythmia data set, the noise being from an MIT-BIH noise pressure data set.
As can be seen from fig. 5, the algorithm of the present invention is good at removing noise while preserving the characteristics of the signal waveform, whether in the face of additive gaussian white noise or mixed noise (including baseline wander, myoelectric noise, motion artifacts). The invention then gains SNR with signal to noise ratio imp And the percentage root mean square difference PRD is used as an index to evaluate the denoising performance of the algorithm, and the denoising performance is compared with other electrocardiosignal denoising algorithms with better performance, wherein the denoising performance comprises sparse representation (Dictionary Learning based Sparse Representation, DLSR) based on dictionary learning, adaptive double-threshold filtering (Discrete Wavelet Transform and Adaptive Dual Threshold Filter, DWT-ADTF) combined with discrete wavelet transformation, EMD-NLM and New edge particle extended Kalman filtering (New Marginalized Particle EKF, new MP-EKF). F in the formula o As the original signal, f n For noisy signals, f d For denoising signals, the noise is additive white gaussian noise.
TABLE 1
As can be seen from the comparison of the performance indexes in Table 1, the invention shows the optimal performance when facing different input signal to noise ratios, which proves that the invention can be better adapted to the varying noise intensity.
In the embodiment, an actual electrocardiograph acquisition device MSP-EXP432P4111 is used as a controller, and ADS1293 is used as an electrocardiograph signal acquisition front-end chip; the acceleration signal is from JY901 nine-axis gesture module. The data is transmitted to a computer through a Bluetooth chip XY-MBD87AD, and the actual denoising effect of the electrocardiosignal with noise is shown as shown in figure 6. The invention adopts MATLAB software to write, and the main program comprises data packet acquisition and unpacking, electrocardiosignal preprocessing, motion state judgment, reference signal recording, multi-threshold SWT denoising, similarity filtering and signal display.
Example two
The invention also provides a self-adaptive electrocardiosignal denoising system based on the reference signal in the static state, which comprises the following steps: the system comprises a signal acquisition module, a judgment module, a distinguishing module, a threshold setting module, a coefficient setting module and a denoising module;
the signal acquisition module is used for acquiring an acceleration signal based on the accelerometer;
the judging module is used for judging the motion state based on the acceleration signal;
the distinguishing module is used for distinguishing the reference signal and the signal to be filtered based on the motion state;
the threshold setting module is used for dynamically setting a threshold by calculating the relation between the reference signal and the signal to be filtered in the multi-threshold SWT denoising stage;
the coefficient setting module is used for dynamically setting coefficients by calculating the relation between the reference signal and the signal to be filtered in the similarity filtering stage;
the denoising module is used for denoising the signal to be filtered based on the threshold value and the coefficient.
The above embodiments are merely illustrative of the preferred embodiments of the present invention, and the scope of the present invention is not limited thereto, but various modifications and improvements made by those skilled in the art to which the present invention pertains are made without departing from the spirit of the present invention, and all modifications and improvements fall within the scope of the present invention as defined in the appended claims.
Claims (8)
1. The self-adaptive electrocardiosignal denoising method based on the reference signal in the static state is characterized by comprising the following steps of:
acquiring an acceleration signal based on the accelerometer;
judging a motion state based on the acceleration signal;
distinguishing a reference signal and a signal to be filtered based on the motion state, wherein the reference signal and the signal to be filtered are electrocardiosignals;
in a multi-threshold SWT denoising stage, dynamically setting a threshold value by calculating the relation between the reference signal and the signal to be filtered;
in the similarity filtering stage, dynamically setting coefficients by calculating the relation between the reference signal and the signal to be filtered;
based on the threshold value and the coefficient, denoising the signal to be filtered is completed;
the multi-threshold SWT denoising method comprises the following steps:
dividing an input signal into a P, T, U wave part with low frequency and a QR and RS wave band part with high frequency;
reserving a high-frequency QR wave band part and a high-frequency RS wave band part;
for the P, T, U wave part with low frequency, 7-level detail coefficient D is obtained through 7-level SWT 1 -D 7 And the final approximation coefficient A, the corresponding wavelet base selects "sym6";
direct removal of detail coefficient D 1 -D 2 The method comprises the steps of carrying out a first treatment on the surface of the Preserving detail coefficient D 6 -D 7 And a final approximation coefficient a; by a predetermined threshold value for detail coefficient D 3 -D 5 Distinguishing;
the similarity filtering method comprises the following steps:
classifying signals by adopting a K-means algorithm, and carrying out non-local mean filtering on the current signal Z (n) to be filtered in each class S:
calculating the weight w (U, Z) of each signal U in the class S relative to the current signal Z (n) to be filtered;
all signals are weighted and summed to obtain a filtered signal
Calculating the energy of the signal to be filtered and the average energy E of the reference signal F A ratio R of (2);
based on the current signal Z (n) to be filtered and after filteringSignal signalAnd the ratio R, the final filtered signal is obtained.
2. The adaptive electrocardiosignal denoising method based on a stationary reference signal according to claim 1, wherein the signal to be filtered is preprocessed before a multi-threshold SWT denoising stage, the preprocessing method comprising: low pass filtering, median filtering and signal segmentation;
wherein, the low pass filtering is: filtering out signals above 80 Hz;
the median filtering is as follows: calculating the median value of the signals to be filtered in each section of window by adopting a sliding window, and subtracting the median value from the signals to be filtered to remove baseline drift;
the signal segmentation is as follows: the signal is segmented by cardiac cycle and resampled to the same length.
3. The adaptive electrocardiosignal denoising method based on a stationary reference signal as claimed in claim 1, wherein the detail coefficient D is determined by a predetermined threshold value 3 -D 5 The distinguishing method comprises the following steps:
dividing the signals into reference signals and signals to be denoised according to the motion state, wherein the set of the reference signals is denoted as Q;
calculating the average standard deviation of the reference signal coefficientsStandard deviation of signal coefficient to be denoised +.>And corresponding ratio r i ；
Wherein the average standard deviationThe expression of (2) is:
standard deviation of the signal coefficient to be denoisedThe expression of (2) is:
the corresponding ratio r i The expression of (2) is:i＝3,4,5。
4. a method of denoising an adaptive electrocardiosignal based on a reference signal at rest as claimed in claim 3 wherein for a signal of length N the threshold corresponds to the expression th i And a thresholding function D i，de The following are provided:
wherein, alpha is used for adjusting the sensitivity degree to noise, and 0.2 is taken.
5. The adaptive electrocardiosignal denoising method based on a reference signal at rest according to claim 1, wherein the filtered signalThe expression of (2) is:
6. the adaptive electrocardiosignal denoising method based on a reference signal at rest according to claim 1, wherein the expression of the ratio R is:
7. the adaptive electrocardiosignal denoising method based on a reference signal at rest according to claim 1, wherein the expression of the final filtered signal is:
8. self-adaptive electrocardiosignal denoising system based on reference signal in static state, which is characterized by comprising: the system comprises a signal acquisition module, a judgment module, a distinguishing module, a threshold setting module, a coefficient setting module and a denoising module;
the signal acquisition module is used for acquiring an acceleration signal based on the accelerometer;
the judging module is used for judging the motion state based on the acceleration signal;
the distinguishing module is used for distinguishing a reference signal and a signal to be filtered based on the motion state;
the threshold setting module is used for dynamically setting a threshold by calculating the relation between the reference signal and the signal to be filtered in a multi-threshold SWT denoising stage;
the coefficient setting module is used for dynamically setting coefficients by calculating the relation between the reference signal and the signal to be filtered in the similarity filtering stage;
the denoising module is used for denoising the signal to be filtered based on the threshold value and the coefficient;
the multi-threshold SWT denoising method comprises the following steps:
dividing an input signal into a P, T, U wave part with low frequency and a QR and RS wave band part with high frequency;
reserving a high-frequency QR wave band part and a high-frequency RS wave band part;
for the P, T, U wave part with low frequency, 7-level detail coefficient D is obtained through 7-level SWT 1 -D 7 And the final approximation coefficient A, the corresponding wavelet base selects "sym6";
direct removal of detail coefficient D 1 -D 2 The method comprises the steps of carrying out a first treatment on the surface of the Preserving detail coefficient D 6 -D 7 And a final approximation coefficient a; by a predetermined threshold value for detail coefficient D 3 -D 5 Distinguishing;
the similarity filtering method comprises the following steps:
classifying signals by adopting a K-means algorithm, and carrying out non-local mean filtering on the current signal Z (n) to be filtered in each class S:
calculating the weight w (U, Z) of each signal U in the class S relative to the current signal Z (n) to be filtered;
all signals are weighted and summed to obtain a filtered signal
Calculating the energy of the signal to be filtered and the average energy E of the reference signal F A ratio R of (2);
based on the current signal Z (n) to be filtered and the filtered signalAnd the ratio R, the final filtered signal is obtained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310385919.6A CN116304777B (en) | 2023-04-12 | 2023-04-12 | Self-adaptive electrocardiosignal denoising method and system based on reference signal during rest |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310385919.6A CN116304777B (en) | 2023-04-12 | 2023-04-12 | Self-adaptive electrocardiosignal denoising method and system based on reference signal during rest |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116304777A CN116304777A (en) | 2023-06-23 |
| CN116304777B true CN116304777B (en) | 2023-11-03 |
Family
ID=86779984
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310385919.6A Active CN116304777B (en) | 2023-04-12 | 2023-04-12 | Self-adaptive electrocardiosignal denoising method and system based on reference signal during rest |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116304777B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8543195B1 (en) * | 2009-11-03 | 2013-09-24 | VivaQuant, LLC | ECG sensing with noise filtering |
| CN104523266A (en) * | 2015-01-07 | 2015-04-22 | 河北大学 | Automatic classification method for electrocardiogram signals |
| CN105769173A (en) * | 2016-02-29 | 2016-07-20 | 浙江铭众科技有限公司 | Electrocardiogram monitoring system with electrocardiosignal denoising function |
| CN108113665A (en) * | 2017-12-14 | 2018-06-05 | 河北大学 | A kind of automatic noise-reduction method of electrocardiosignal |
| CN109359506A (en) * | 2018-08-24 | 2019-02-19 | 浙江工业大学 | A kind of mcg-signals noise-reduction method based on wavelet transformation |
| CN110051325A (en) * | 2019-03-29 | 2019-07-26 | 重庆邮电大学 | Electrocardiosignal integrated filter method based on wavelet transformation and improvement EEMD |
| CN112603325A (en) * | 2020-12-11 | 2021-04-06 | 上海交通大学 | Electrocardio abnormity discrimination system and method based on morphological filtering and wavelet threshold |
| CN113079538A (en) * | 2020-01-03 | 2021-07-06 | 中国科学院大学 | Heterogeneous equipment cooperative transmission mechanism based on cross-technology communication technology |
| CN115836846A (en) * | 2022-12-14 | 2023-03-24 | 北京航空航天大学 | Non-invasive blood pressure estimation method based on self-supervision transfer learning |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI780884B (en) * | 2021-08-31 | 2022-10-11 | 國立中正大學 | Single image deraining method and system thereof |
-
2023
- 2023-04-12 CN CN202310385919.6A patent/CN116304777B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8543195B1 (en) * | 2009-11-03 | 2013-09-24 | VivaQuant, LLC | ECG sensing with noise filtering |
| CN104523266A (en) * | 2015-01-07 | 2015-04-22 | 河北大学 | Automatic classification method for electrocardiogram signals |
| CN105769173A (en) * | 2016-02-29 | 2016-07-20 | 浙江铭众科技有限公司 | Electrocardiogram monitoring system with electrocardiosignal denoising function |
| CN108113665A (en) * | 2017-12-14 | 2018-06-05 | 河北大学 | A kind of automatic noise-reduction method of electrocardiosignal |
| CN109359506A (en) * | 2018-08-24 | 2019-02-19 | 浙江工业大学 | A kind of mcg-signals noise-reduction method based on wavelet transformation |
| CN110051325A (en) * | 2019-03-29 | 2019-07-26 | 重庆邮电大学 | Electrocardiosignal integrated filter method based on wavelet transformation and improvement EEMD |
| CN113079538A (en) * | 2020-01-03 | 2021-07-06 | 中国科学院大学 | Heterogeneous equipment cooperative transmission mechanism based on cross-technology communication technology |
| CN112603325A (en) * | 2020-12-11 | 2021-04-06 | 上海交通大学 | Electrocardio abnormity discrimination system and method based on morphological filtering and wavelet threshold |
| CN115836846A (en) * | 2022-12-14 | 2023-03-24 | 北京航空航天大学 | Non-invasive blood pressure estimation method based on self-supervision transfer learning |
Non-Patent Citations (2)
| Title |
|---|
| "Electrocardiogram Signal Denoising Based on Multi-Threshold Stationary Wavelet Transform";Huyang Peng.et al;《IEEE》;全文 * |
| "基于无迹卡尔曼滤波估计的无线传感器网络时钟分辨率优化";何灏等;《电子与信息学报》;第41卷(第3期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116304777A (en) | 2023-06-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105919584B (en) | Heart rate method of estimation and device for wearable heart rate monitor apparatus | |
| CN110916636B (en) | BCG signal heart rate calculation method and system based on dynamic second-order differential threshold | |
| KR20140139564A (en) | Systems and methods for ecg monitoring | |
| CN107361764B (en) | Method for rapidly extracting electrocardiosignal characteristic waveform R wave | |
| Li et al. | Application of an EMG interference filtering method to dynamic ECGs based on an adaptive wavelet-Wiener filter and adaptive moving average filter | |
| Seljuq et al. | Selection of an optimal mother wavelet basis function for ECG signal denoising | |
| CA3170821A1 (en) | Fusion signal processing for maternal uterine activity detection | |
| CN112587133A (en) | Method for measuring blood oxygen saturation | |
| CN110507317B (en) | Self-adaptive CA-CFAR (Carrier frequency-constant false alarm) positioning method for electrocardiosignal R wave | |
| Janušauskas et al. | Ensemble empirical mode decomposition based feature enhancement of cardio signals | |
| Nguyen et al. | Artifact elimination in ECG signal using wavelet transform | |
| CN111956209B (en) | Electrocardiosignal R wave identification method based on EWT and structural feature extraction | |
| CN116304777B (en) | Self-adaptive electrocardiosignal denoising method and system based on reference signal during rest | |
| Ihza et al. | Study of Denoising Method to Detect Valvular Heart Disease Using Phonocardiogram (PCG) | |
| Li et al. | A High-Efficiency and Real-Time Method for Quality Evaluation of PPG Signals | |
| CN113229826B (en) | QRS wave detection method and device and electronic equipment | |
| CN113925482A (en) | Heart rate calculation method, wearable electronic device and storage medium | |
| Reddy et al. | A Fast Iterative Filtering Method for Efficient Denoising of Phonocardiogram Signals | |
| Georgieva-Tsaneva | A novel photoplethysmographic noise removal method via wavelet transform to effective preprocessing | |
| Shao et al. | A Photoplethysmograph Signal Preprocess Method Based on Wavelet Transform | |
| CN112401924B (en) | Heart sound segmentation method and device | |
| Das et al. | On an algorithm for detection of QRS complexes in noisy electrocardiogram signal | |
| CN117426776B (en) | Electrocardiogram signal characteristic intelligent extraction method | |
| Hu et al. | A robust beat-to-beat artifact detection algorithm for pulse wave | |
| Peng et al. | R-peak Detection for ECG Biomedical Monitoring |
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 |