Improved wavelet threshold electrocardiosignal denoising method

By improving the wavelet threshold method, smooth decomposition and wavelet transformation of the ECG signal, combined with mathematical morphological processing, the problem of modal aliasing and computing resource occupancy in ECG signal denoising is solved, efficient denoising is achieved, and the Signal signal-to-noise ratio and diagnostic accuracy.

CN118924306APending Publication Date: 2024-11-12CHANGCHUN UNIV OF SCI & TECH +2
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410983701.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-22
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

The prior art is prone to modal aliasing and signal distortion in the process of denoising electrocardiogram signals, and the computing resources are large, making it difficult to effectively deal with high-frequency and low-frequency noises, affecting the accuracy of disease diagnosis.

Method used

The improved wavelet threshold method is adopted to denoise the ECG signal through smooth decomposition and wavelet transformation, and the grouping threshold is determined using the noise-energy ratio and the ECG average period, and the appropriate wavelet basis and decomposition layers are selected for wavelet threshold denoising, and The final denoising signal is reconstructed by mathematical morphology.

Benefits of technology

It effectively reduces high-frequency and low-frequency noise interference, improves the signal-to-noise ratio of the signal to the noise, maintains the integrity and accuracy of the ECG waveform, improves the calculation efficiency, and reduces signal distortion and noise residue.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118924306A_ABST
    Figure CN118924306A_ABST
Patent Text Reader

Abstract

The invention belongs to the technical field of computers, and particularly relates to an improved wavelet threshold electrocardiosignal denoising method, which comprises the following steps of: decomposing electrocardiosignals by using smooth decomposition, and denoising noise component groups of the electrocardiosignals by using an improved wavelet threshold algorithm and mathematical morphology; the method comprises the following steps: firstly, decomposing a noisy electrocardiosignal into a plurality of smooth decomposition components from high frequency to low frequency, establishing a boundary threshold, and dividing the smooth decomposition components into three groups: a high-frequency noise component group, a useful signal component group and a low-frequency noise component group; secondly, denoising processing is carried out on a component group containing myoelectricity noise and other high-frequency noise by using an improved wavelet threshold denoising method, and denoising processing is carried out on a component group containing baseline drift and other low-frequency noise by using mathematical morphology; and finally, reconstructing each component group to obtain a final de-noised electrocardiosignal.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the field of computer technology, and in particular is an improved wavelet threshold electrocardiogram signal denoising method. Background Art

[0002] ECG signals are manifestations of cardiac activity on the human body surface and are an important basis for judging various arrhythmias, myocardial infarction and other diseases. The amplitude range of ECG signals is between 10uV and 5mV. It is a non-stationary, nonlinear and very weak physiological signal. It contains both noise signals and useful signals, which is of great significance for the prevention and treatment of cardiovascular diseases. Due to the simplicity and non-invasiveness of ECG signals, doctors often use it to diagnose heart disease. Useful signals mainly include QRS waves (indicating ventricular depolarization), P waves (indicating atrial depolarization) and T waves (indicating ventricular repolarization). Noise signals include high-frequency noise (such as myoelectric noise, spinal cord interference, electrical interference, etc.) and low-frequency noise (such as baseline drift, etc.). In the process of collecting ECG signals, high-frequency noise will affect the amplitude and frequency of the R wave, thereby interfering with the diagnosis of diseases such as arrhythmias. Low-frequency noise can cause ST segment shift, resulting in the loss of P and T wave peaks, affecting the diagnosis of diseases such as myocardial ischemia and ventricular hypertrophy. Therefore, in order to improve the accuracy of disease diagnosis, it is very necessary to reduce the high-frequency and low-frequency noise in the ECG signal.

[0003] Due to the nonlinear and non-stationary characteristics of ECG signals, Fourier analysis is difficult to effectively solve the problem of spectrum changing over time. Neural network learning methods require a large number of example signals for training. However, the ECG signals of subjects at different times and spaces vary greatly, resulting in poor versatility of training results. Empirical mode decomposition (EMD) is adaptive and more conducive to ECG signal denoising. However, EMD has mode aliasing in the process of ECG signal decomposition, which has a certain impact on the success rate of signal analysis. Although the comprehensive empirical mode decomposition (EEMD) solves the mode aliasing problem of EMD, the increase in the total number leads to an exponential increase in the amount of calculation for repeated decomposition. In addition, the introduction of auxiliary noise brings about the problems of noise residue and signal reconstruction error. The complete set empirical mode decomposition with adaptive noise (CEEMDAN) requires a lot of computing resources when processing long-term ECG signals, and is sensitive to the non-stationarity of the signal, which easily causes signal distortion. There are inevitable defects in using soft thresholds and hard thresholds to process ECG signals. Although the signal processed by the soft threshold method is relatively smooth, the signal amplitude will decay, especially the decay of the r wave amplitude is more obvious, the signal edge will become blurred, resulting in signal distortion. Although the hard threshold method can reconstruct the feature points of the original signal well, the signal processed by the hard threshold method will oscillate and show pseudo-Gibbs phenomenon.

[0004] Analyzing ECG signals is an important means to study cardiovascular diseases, and ECG signal denoising methods have emerged in an endless stream. The technical problems of wavelet threshold denoising, empirical mode decomposition denoising, and adaptive noise complete set empirical mode decomposition denoising are easy to produce modal aliasing and signal distortion and occupy a lot of computing resources. Therefore, an improved wavelet threshold ECG signal denoising method is proposed to solve the above problems. Summary of the invention

[0005] In order to make up for the deficiencies of the prior art, at least one technical problem raised in the background technology is solved.

[0006] The technical solution adopted by the present invention to solve the technical problem is: the improved wavelet threshold electrocardiogram signal denoising method described in the present invention comprises the following steps:

[0007] S1. Add the myoelectric noise and baseline drift noise from the MIT-BIH noise stress test database to the ECG signal to provide a noisy ECG signal for the denoising algorithm.

[0008] S2. Smoothing decomposition (sdc) of noisy ECG signals makes the analysis of ECG signals more accurate.

[0009] S3, determine the grouping threshold of the smooth decomposition component by the noise energy ratio and the average ECG cycle,

[0010] That is, the smooth decomposition high-frequency component group threshold (HBth) and the smooth decomposition low-frequency component group threshold (HBth). The high-frequency threshold and the low-frequency threshold are used to divide the smooth decomposition components into high-frequency noise component group and low-frequency noise component group, and the remaining components are the useful signal component group (sdc u ).

[0011] S4, determine the wavelet threshold of each component function according to the average frequency of the smooth decomposition component, that is,

[0012] λ i =[median(f i ) / 0.6745]·X

[0013] According to the sampling frequency (F s ) and the time width of the useful signal (T W ) determines the structural element (D), that is,

[0014] D=F S *T W

[0015] Where: is the frequency of the smoothed decomposition component of the ECG signal, and X is a constant;

[0016] S5, by selecting the appropriate wavelet basis and decomposition layer number for the high-frequency noise component group, a set of wavelet decomposition coefficients is obtained, and the improved wavelet threshold denoising is performed to obtain the denoised high-frequency component group (sdc whd ); The low-frequency noise component group is denoised by mathematical morphology. This method uses a linear structural element to remove low-frequency noise and obtains a denoised low-frequency component group (sdc mm ).

[0017] S6, reconstruct the high-frequency noise component group processed by wavelet threshold denoising and the low-frequency noise component group processed by mathematical morphology with the useful noise component group, that is,

[0018]

[0019] Where: N is the number of smooth decomposition layers;

[0020] The final denoised signal is obtained by reconstructing the denoised high-frequency noise component, low-frequency noise component and useful signal component.

[0021] Preferably, the improved wavelet threshold function in step 5 is expressed as:

[0022]

[0023] Among them, C i,k The kth coefficient of the ith component.

[0024] The beneficial effects of the present invention are:

[0025] 1. The present invention provides an improved wavelet threshold ECG signal denoising method, which effectively solves the main high-frequency noise problem of electromyographic noise and the main low-frequency noise problem of baseline drift, while improving the signal-to-noise ratio of the signal and maintaining the integrity and accuracy of the ECG waveform.

[0026] 2. The present invention provides an improved wavelet threshold ECG signal denoising method. The denoising effect of wavelet decomposition depends on the selection of wavelet function, the empirical mode decomposition has mode aliasing, and the calculation amount of adaptive noise complete set empirical mode decomposition is too large. Smoothing decomposition is used to decompose the ECG signal, and the improved wavelet threshold is used for high-frequency denoising, thereby improving the calculation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0028] Figure 1 is a flow chart of the method of the present invention;

[0029] Figure 2 This is a comparison chart of specific applications of the method of the present invention. DETAILED DESCRIPTION

[0030] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0031] See also Figure 1-2 As shown, an improved wavelet threshold ECG signal denoising method comprises the following steps:

[0032] S1, adding myoelectric noise and baseline drift noise from the MIT-BIH noise stress test database to the ECG signal to provide a noisy ECG signal for the denoising algorithm;

[0033] S2, smoothing and decomposing the noisy ECG signal, making the analysis of the ECG signal more accurate;

[0034] S3, determining the grouping threshold of the smooth decomposition components by the noise energy ratio and the ECG average cycle, and dividing the smooth decomposition components into a high-frequency noise component group, a useful signal component group, and a low-frequency noise component group;

[0035] S4, determining a threshold value of each component function according to the average frequency of the smoothed decomposition component;

[0036] Determine the structural element according to the sampling frequency and the time width of the useful signal;

[0037] S5. By selecting a suitable wavelet basis and decomposition layer number for the high-frequency noise component group, a set of wavelet decomposition coefficients is obtained, and improved wavelet threshold denoising is performed. The low-frequency noise component group is denoised by mathematical morphology. This method uses linear structural elements to remove low-frequency noise;

[0038] S6. Reconstruct the high-frequency noise component group processed by wavelet threshold denoising and the low-frequency noise component group processed by mathematical morphology with the useful noise component group to obtain a final denoised signal.

[0039] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.

Claims

1. An improved wavelet threshold ECG signal denoising method, characterized by: An improved wavelet threshold ECG signal denoising method comprises the following steps: S1, adding myoelectric noise and baseline drift noise from the MIT-BIH noise stress test database to the ECG signal to provide a noisy ECG signal for the denoising algorithm; S2, smoothing decomposition (sdc) of noisy ECG signals to make the analysis of ECG signals more accurate; S3, determine the grouping threshold of the smooth decomposition components by the noise energy ratio and the ECG average cycle, that is, the smooth decomposition high-frequency component group threshold (HBth) and the smooth decomposition low-frequency component group threshold (HBth), and divide the smooth decomposition components into a high-frequency noise component group and a low-frequency noise component group by the high-frequency threshold and the low-frequency threshold. The remaining components are the useful signal component group (sdc u ); S4. Determine the wavelet threshold of each component function according to the average frequency of the smooth decomposition component, that is: l i =[median(f i ) / 0.6745]·X According to the sampling frequency (F s ) and the time width of the useful signal (T W ) determines the structural element (D), namely: D=F S *T W Where: f i is the frequency of the smoothed decomposed component of the ECG signal, and X is a constant; S5, by selecting the appropriate wavelet basis and decomposition layer number for the high-frequency noise component group, a set of wavelet decomposition coefficients is obtained, and the improved wavelet threshold denoising is performed to obtain the denoised high-frequency component group (sdc whd ); The low-frequency noise component group is denoised by mathematical morphology. This method uses a linear structural element to remove low-frequency noise and obtains a denoised low-frequency component group (sdc mm ); S6, reconstruct the high-frequency noise component group processed by wavelet threshold denoising and the low-frequency noise component group processed by mathematical morphology with the useful noise component group, that is: Where: N is the number of smooth decomposition layers; The final denoised signal is obtained by reconstructing the denoised high-frequency noise component, low-frequency noise component and useful signal component.

2. The improved wavelet threshold ECG signal denoising method according to claim 1, characterized in that: The improved wavelet threshold function in S5 is expressed as: Among them, C i,k The kth coefficient of the i-th wavelet decomposition component.