CN116671932A - Depression brain electric signal extraction method based on wavelet and self-adaptive filtering - Google Patents
Depression brain electric signal extraction method based on wavelet and self-adaptive filtering Download PDFInfo
- Publication number
- CN116671932A CN116671932A CN202310660335.5A CN202310660335A CN116671932A CN 116671932 A CN116671932 A CN 116671932A CN 202310660335 A CN202310660335 A CN 202310660335A CN 116671932 A CN116671932 A CN 116671932A
- Authority
- CN
- China
- Prior art keywords
- wavelet
- signal
- signals
- depression
- 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.)
- Pending
Links
- 238000001914 filtration Methods 0.000 title claims abstract description 56
- 238000000605 extraction Methods 0.000 title claims abstract description 16
- 210000004556 brain Anatomy 0.000 title claims description 29
- 230000003044 adaptive effect Effects 0.000 claims abstract description 42
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 29
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 25
- 238000000034 method Methods 0.000 claims abstract description 19
- 238000004458 analytical method Methods 0.000 claims abstract description 9
- 238000007781 pre-processing Methods 0.000 claims abstract description 6
- 238000012545 processing Methods 0.000 claims description 16
- 238000005070 sampling Methods 0.000 claims description 12
- 230000009466 transformation Effects 0.000 claims description 9
- 230000000694 effects Effects 0.000 abstract description 6
- 238000005516 engineering process Methods 0.000 abstract description 4
- 238000003745 diagnosis Methods 0.000 abstract description 3
- 230000007547 defect Effects 0.000 abstract description 2
- 230000001737 promoting effect Effects 0.000 abstract 2
- 230000003183 myoelectrical effect Effects 0.000 abstract 1
- 230000006870 function Effects 0.000 description 10
- 230000008569 process Effects 0.000 description 4
- 208000020401 Depressive disease Diseases 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 201000010099 disease Diseases 0.000 description 3
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 3
- 230000006698 induction Effects 0.000 description 3
- 208000020016 psychiatric disease Diseases 0.000 description 3
- 230000009467 reduction Effects 0.000 description 3
- 230000004044 response Effects 0.000 description 3
- 230000007958 sleep Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 230000005611 electricity Effects 0.000 description 2
- 230000008451 emotion Effects 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 239000003550 marker Substances 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000002085 persistent effect Effects 0.000 description 2
- 230000000306 recurrent effect Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000033764 rhythmic process Effects 0.000 description 2
- 206010010144 Completed suicide Diseases 0.000 description 1
- 241001664469 Tibicina haematodes Species 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 239000000654 additive Substances 0.000 description 1
- 230000000996 additive effect Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000005311 autocorrelation function Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000000133 brain stem Anatomy 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 210000003710 cerebral cortex Anatomy 0.000 description 1
- 230000001684 chronic effect Effects 0.000 description 1
- 230000019771 cognition Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000002405 diagnostic procedure Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000009533 lab test Methods 0.000 description 1
- 230000003340 mental effect Effects 0.000 description 1
- 230000036651 mood Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 229940001470 psychoactive drug Drugs 0.000 description 1
- 239000004089 psychotropic agent Substances 0.000 description 1
- 230000005180 public health Effects 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000037322 slow-wave sleep Effects 0.000 description 1
- 230000004936 stimulating effect Effects 0.000 description 1
- 230000000638 stimulation Effects 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/16—Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state
- A61B5/165—Evaluating the state of mind, e.g. depression, anxiety
-
- 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
- A61B5/369—Electroencephalography [EEG]
-
- 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
- A61B5/369—Electroencephalography [EEG]
- A61B5/372—Analysis of electroencephalograms
-
- 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
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/725—Details of waveform analysis using specific filters therefor, e.g. Kalman or adaptive filters
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7253—Details of waveform analysis characterised by using transforms
- A61B5/726—Details of waveform analysis characterised by using transforms using Wavelet transforms
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Psychiatry (AREA)
- Public Health (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- Pathology (AREA)
- Biophysics (AREA)
- Medical Informatics (AREA)
- Veterinary Medicine (AREA)
- General Health & Medical Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Surgery (AREA)
- Heart & Thoracic Surgery (AREA)
- Signal Processing (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physiology (AREA)
- Psychology (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Developmental Disabilities (AREA)
- Hospice & Palliative Care (AREA)
- Algebra (AREA)
- Software Systems (AREA)
- Child & Adolescent Psychology (AREA)
- Computational Mathematics (AREA)
- Educational Technology (AREA)
- General Engineering & Computer Science (AREA)
- Social Psychology (AREA)
- Databases & Information Systems (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The invention discloses a depression electroencephalogram signal extraction method based on wavelet and self-adaptive filtering, which aims at the defects that the extraction effect is not ideal under the condition of low signal-to-noise ratio of the adaptive filtering, the extraction effect is good when the signal-to-noise ratio is high, and the extraction effect is poor when the signal-to-noise ratio is low, and the wavelet analysis method is used for preprocessing an input signal, filtering noise signals such as electrooculogram, electrocardiograph, myoelectric and the like, and improving the signal-to-noise ratio of the noise-containing signal; after wavelet decomposition is carried out on the extracted electroencephalogram signals, an adaptive filtering algorithm is introduced into wavelet coefficients of each layer for filtering; the electroencephalogram signals are obtained through wavelet coefficient reconstruction, the electroencephalogram signals extracted by the method are clear and neat, compared with the adaptive filtering algorithm, the convergence step length of the algorithm is increased, the convergence speed and stability of the LMS algorithm are improved, and noise is well suppressed; the invention is helpful for improving the diagnosis and treatment effect of depression, promoting the progress of brain-computer interface technology and promoting the development of brain-computer interface technology.
Description
Technical Field
The invention belongs to the technical field of bioelectric signal processing, and particularly relates to a depression electroencephalogram signal extraction method based on wavelet and adaptive filtering.
Background
Depression is a common, chronic recurrent disease, manifested by a marked and persistent depression in the mood, accompanied by corresponding changes in thinking and behavior, with a tendency to recurrent attacks, normal intermittent mental functioning, without residual personality defects. At present, about 5% of people in China have psychological disorders with different degrees, about 15% of depression patients die from suicide, and in recent years, the tendency is rising. Depressive disorders have become an increasingly serious public health problem in our country. A joint study by the world health organization, world banks and Harvard university has shown that depression has become the second most serious disease burden in China. Experts speculate that the 21 st century is the peak in the onset of depression. However, most patients with depression are not correctly diagnosed and treated due to the lack of knowledge of the disease in the past. With the current state of research and development, patients with depression have no positive results from specific laboratory tests. Therefore, the current depression diagnosis method is to perform comprehensive analysis and evaluation on various data (cognition, emotion, personality, social functions and the like) obtained from various information sources (patient, relatives and outpatients) by various data collection methods (interviews, behavioral observations, psychological tests, laboratory physiological examinations and the like) so as to make the most reliable diagnosis and evaluation of symptom severity. However, such time consuming, cumbersome and uneconomical diagnostic procedures are difficult to use in practical clinical practice.
Electroencephalogram (EEG) is an electrical activity signal of nerve cells recorded through the cerebral cortex, and has many kinds of brain wave rhythms and various changes. The heart state affects the changes of brain waves in various emotions. Research shows that the brain electrical signals of depression patients show abnormal phenomena in parameters such as rhythms, amplitude, power and the like. The united states Maixner recognizes that slow wave sleep, particularly reduced total sleep, may be a characteristic of depression and other mental disorders, and is more stable in sleep monitored overnight, presumably a marker of diathesis. Lauer found that this reduction in total sleep volume was associated with ventricle size. There are many studies supporting that MMN reduction is a predisposition marker for depression, indicating that persistent MMN reduction in depression is independent of psychotropic drugs and also occurs in the first-degree relatives of depression. These features make it believed that the EEG researchers must have a great deal of information available to us in EEG signals. Thus, intensive research into EEG has been significant in diagnosing mental disorders.
Disclosure of Invention
Aiming at the technical requirements, the invention discloses a depression electroencephalogram signal extraction method based on wavelet and adaptive filtering, which can improve the convergence rate of the adaptive filtering, improve the accuracy of acquiring depression electroencephalogram signals and forcefully inhibit the noise of signals.
The purpose of the invention is realized in the following way:
a depression electroencephalogram signal extraction method based on wavelet and adaptive filtering comprises the following steps:
step a, acquiring brain electrical signal data of a patient suffering from depression;
step b, preprocessing the signals by using a wavelet analysis method, selecting a proper wavelet basis function to carry out wavelet decomposition, filtering out useless physiological electric signals such as electrooculogram, myoelectricity, electrocardiograph and the like, and improving the signal-to-noise ratio of the signals;
step c, extracting the electroencephalogram signals by a self-adaptive filtering algorithm, and performing secondary wavelet transformation on the electroencephalogram signals;
step d, taking the reconstructed noise as a reference input of the adaptive filter, and performing adaptive filtering processing to obtain a denoising signal;
and e, reconstructing the signals obtained in the step d and the brain electrical signals of each frequency band in the step b to obtain the effective brain electrical signals of the patient suffering from depression.
According to the depression electroencephalogram signal extraction method based on wavelet and adaptive filtering, after the electroencephalogram signal data are collected, amplified and filtered through the signal collecting unit, the electroencephalogram signal data are converted into digital signals through the A/D converter and are sent to the real-time electroencephalogram signal processing module, and the real-time electroencephalogram signal processing module is used for eliminating noise caused by interference and physiological artifacts.
Step b, realizing wavelet decomposition by adopting a Mallat algorithm, and decomposing signals into wavelet coefficients with different scales, wherein the basic formula of the Mallat algorithm is as follows:
where j is the scale of the wavelet transform, k is the index corresponding to the scale factor or wavelet coefficient, w j,k For the kth wavelet coefficient in the jth layer wavelet decomposition, n represents the position in the time domain, x j+1,2k-n For the 2k-n low frequency coefficients in the j+1 layer wavelet decomposition, y j+1,2k-n In order to obtain the 2k-n high frequency coefficients in the j+1 layer wavelet decomposition, h (n) is the low pass filter coefficient at the n-th sampling point in the wavelet decomposition, and g (n) is the high pass filter coefficient at the n-th sampling point in the wavelet decomposition.
For the detail coefficient of each scale, low-pass filtering or high-pass filtering is realized by setting a threshold value so as to remove noise signals.
Wavelet reconstruction is carried out on the processed approximation coefficient and detail coefficient to obtain a denoised signal, and the basic formula of the wavelet reconstruction is as follows:
wherein x is n N is the position in time domain, J is the scale of wavelet transformation, k is the index corresponding to scale coefficient or wavelet coefficient, J is the highest number of layers of decomposition,for the low-pass filter coefficients at the moment of the j-th layer k +.>For the high-pass filter coefficient at the j-th layer k moment, a j,k,n For the approximation coefficient of the jth layer of the nth sampling point, d j,k,n Is the detail coefficient of the jth layer at the nth sampling point.
The beneficial effects are that:
the invention relates to a depression electroencephalogram signal extraction method based on wavelet and self-adaptive filtering. After wavelet decomposition is carried out on the extracted electroencephalogram signals, an adaptive filtering algorithm is introduced into each layer of wavelet coefficients for filtering, and then the electroencephalogram signals are obtained through wavelet coefficient reconstruction, so that the electroencephalogram signals extracted by the method are clear and neat, compared with the adaptive filtering algorithm, the convergence step length of the algorithm is increased, the convergence speed and stability of the LMS algorithm are improved, noise is better suppressed, and more accurate and stable electroencephalogram signals can be provided for brain-computer interface technology; in addition, the brain-computer interface technology is used, so that the brain-computer signal of a person can be connected with external equipment, and man-machine interaction is realized.
Drawings
FIG. 1 is a flow chart of a method for extracting an electrical signal of a depressive disorder brain based on wavelet and adaptive filtering of the present invention;
FIG. 2 is a different signal-to-noise ratio signal extraction;
fig. 3 is a signal waveform of wavelet denoising in combination with adaptive filtering.
Detailed Description
The following describes the embodiments of the present invention in further detail with reference to the accompanying drawings.
Detailed description of the preferred embodiments
The depression electroencephalogram signal extraction method based on wavelet and adaptive filtering in the specific embodiment comprises the following steps of:
step a, acquiring brain electrical signal data of a patient suffering from depression;
step b, preprocessing the signals by using a wavelet analysis method, selecting a proper wavelet basis function to carry out wavelet decomposition, filtering out useless physiological electric signals such as electrooculogram, myoelectricity, electrocardiograph and the like, and improving the signal-to-noise ratio of the signals;
step c, extracting the electroencephalogram signals by a self-adaptive filtering algorithm, and performing secondary wavelet transformation on the electroencephalogram signals;
step d, taking the reconstructed noise as a reference input of the adaptive filter, and performing adaptive filtering processing to obtain a denoising signal;
and e, reconstructing the signals obtained in the step d and the brain electrical signals of each frequency band in the step b to obtain the effective brain electrical signals of the patient suffering from depression.
According to the depression electroencephalogram signal extraction method based on wavelet and adaptive filtering, after the electroencephalogram signal data are collected, amplified and filtered through the signal collecting unit, the electroencephalogram signal data are converted into digital signals through the A/D converter and are sent to the real-time electroencephalogram signal processing module, and the real-time electroencephalogram signal processing module is used for eliminating noise caused by interference and physiological artifacts.
Step b, realizing wavelet decomposition by adopting a Mallat algorithm, and decomposing signals into wavelet coefficients with different scales, wherein the basic formula of the Mallat algorithm is as follows:
where j is the scale of the wavelet transform, k is the index corresponding to the scale factor or wavelet coefficient, w j,k For the kth wavelet coefficient in the jth layer wavelet decomposition, n represents the position in the time domain, x j+1,2k-n For the 2k-n low frequency coefficients in the j+1 layer wavelet decomposition, y j+1,2k-n In order to obtain the 2k-n high frequency coefficients in the j+1 layer wavelet decomposition, h (n) is the low pass filter coefficient at the n-th sampling point in the wavelet decomposition, and g (n) is the high pass filter coefficient at the n-th sampling point in the wavelet decomposition.
For the detail coefficient of each scale, low-pass filtering or high-pass filtering is realized by setting a threshold value so as to remove noise signals.
Wavelet reconstruction is carried out on the processed approximation coefficient and detail coefficient to obtain a denoised signal, and the basic formula of the wavelet reconstruction is as follows:
wherein x is n N is the position in time domain, J is the scale of wavelet transformation, k is the index corresponding to scale coefficient or wavelet coefficient, J is the highest number of layers of decomposition,for the low-pass filter coefficients at the moment of the j-th layer k +.>For the high-pass filter coefficient at the j-th layer k moment, a j,k,n For the approximation coefficient of the jth layer of the nth sampling point, d j,k,n Is the detail coefficient of the jth layer at the nth sampling point.
Detailed description of the preferred embodiments
The method for extracting the depression electroencephalogram signal based on the wavelet and the adaptive filtering in the specific embodiment expands on the basis of the first specific embodiment and comprises the following steps of:
step a, acquiring brain electrical signal data of a patient suffering from depression;
extracting EEG signals by using an EEG signal acquisition system; the specific implementation mode is as follows: an EEG brain electricity acquisition system Neuroscan which is one of the tools for acquiring and analyzing brain electricity data with the largest internationally-used range and extremely strong professional is used for acquisition; it mainly includes electrode cap, power supply box, head box, control box, collecting and analyzing tool and stimulating tool. In the acquisition process, synAmps is an amplifier adopted in a Neuroscan electroencephalogram acquisition and analysis system. After the processing of the head box amplifier, the acquired original signals can be amplified, the digital conversion of the signals is completed, the SynAmps amplifier can ensure that the distortion of the signals is reduced to the minimum, and the repeated stimulation is carried out 4000 times in the process of acquiring the EEG signals for subsequent processing.
Step b, preprocessing the signals by using a wavelet analysis method, selecting a proper wavelet basis function to carry out wavelet decomposition, filtering out useless physiological electric signals such as electrooculogram, myoelectricity, electrocardiograph and the like, and improving the signal-to-noise ratio of the signals;
and selecting a proper wavelet base, preprocessing an input signal by using a wavelet analysis method, filtering useless interference signals such as electrooculogram, myoelectricity, electrocardiograph and the like, and improving the specific rule of the signal-to-noise ratio of the signal. Aiming at the time domain characteristics of the depression brain electrical signals, the sound induction brain electrical signals are effectively extracted in the time domain, and the induction signals can be accurately extracted by selecting wavelet basis functions in combination with the clinical characteristics of the hearing induction brain stem reaction signals. These characteristics are optimized for the signal to be extracted based on the characteristics of the wavelet basis function, such as tight support, regularization, orthogonality and double crossover, symmetry, vanishing distance.
Step c, extracting the electroencephalogram signals by a self-adaptive filtering algorithm, and performing secondary wavelet transformation on the electroencephalogram signals;
step d, taking the reconstructed noise as a reference input of the adaptive filter, and performing adaptive filtering processing to obtain a denoising signal;
step e, reconstructing the signals obtained in the step d and the brain electrical signals of each frequency band in the step b to obtain effective brain electrical signals of the patient suffering from depression;
mixed signal f (t) =s (t) +d i (t), s (t) is the original brain electrical signal, d i (t) is a noise signal mixed in an electroencephalogram signal, mainly comprising some noises caused by electrooculogram, myoelectricity and the like, and recovering a real signal from the undistorted f (t), wherein the step of adopting a wavelet threshold to remove noise is as follows:
(1) Selecting proper wavelet base phi (t) and corresponding decomposition scale j for carrying out binary discrete wavelet transformation on the signal containing noise to obtain wavelet coefficient w on each scale i,k 。
(2) And processing the wavelet coefficient by adopting a certain threshold t and a corresponding threshold processing function to obtain a processed coefficient.
(3) For the processed coefficient w i,k Performing small sizeWave reconstruction yields an estimated signal for the true signal s.
The specific principle of denoising and filtering the depression brain electrical signal by using adaptive filtering and analyzing and combining wavelet is as follows:
input signal vector: x (n) = [ X (n), X (n-1), …, X (n-m+1)] T
Input signal weight coefficient vector: w (n) = [ W (n), W (n-1), …, W (n-m+1)] T
The output signal y (n) is the product of the input vector X (n) and the weight coefficient vector W (n):
filter output error: e (n) =d (n) -W T (n)X(n)
The weight coefficient vector of the current moment is output through the step factor mu and the error of the previous moment, so that E [ E ] 2 (n)]At a minimum, the output error only preserves the portion that is uncorrelated with x (n): e (E (n) x (n-M))=0, 0.ltoreq.m.ltoreq.M-1
The weight coefficient vector at the current moment obtained by using the step factor, the weight coefficient vector obtained at the previous moment and the output error can be expressed as follows: w (n+1) =w (n) -2 μe (n) X (n)
The mean square value of the output error is shown as:
fig. 1 is a flowchart of a method for extracting a wavelet and adaptively filtered depression brain electrical signal.
Fig. 2 is a graph of signal extraction waveforms for different signal to noise ratios, resulting in a clean waveform by increasing the signal to noise ratio.
The unknown channel is in a Finite Impulse Response (FIR) structure, an adaptive filter in the FIR structure is constructed, a pseudo-random series is used as an input signal x (n) of the system, and the input signal x (n) is simultaneously sent into the unknown channel system and the adaptive filter. The coefficients of the adaptive filter are adjusted such that the mean square error of the error signal e (n) is minimized and the output y (n) of the adaptive filter is approximately equal to the output d (n) of the communication system. It can be demonstrated that the presence of additive noise v (n) does not affect the final convergence of the adaptive filter to the optimal wiener solution.
Two FIR systems with the same input and similar output should have similar characteristics. Thus, the characteristics of the adaptive filter or its unit impulse response may be employed to approximate the characteristics or unit impulse response of the replacement unknown system. The adaptive filter consists of two parts, one part being a digital filter for performing the desired signal processing and the other part being an adaptive algorithm for adjusting the filter coefficients (or weights). The minimum mean square error criterion four minimizes the mean square value of the output e (k) of the canceller, namely: e (k) =e [ E ] 2 (k)]。
e (k) is called a criterion function, the calculation process of the adaptive filtering algorithm is as follows, and then the output z (k) of the filter is calculated at the moment k; calculating e (k) =y (k) -z (k); calculating the next filter parameter according to the principle of minimizing the criterion function e; k=k+1, and jump to (1) gradually iterate E [ E ] 2 (k)]Approximation es 2 (k)]This is to make e (k) successive to s (k), and when the algorithm converges, the measured signal s (k) with noise removed is obtained from the output of the canceller.
The input signal x (n) is passed through a digital filter with adjustable parameters to produce an output signal y (n) which is compared with a reference signal d (n) to form an error signal e (n). In design, knowledge about the statistical properties of the input signal and noise is not required in advance, and the required statistical properties can be gradually known or estimated in the working process of the user. And based on the change, the method automatically adjusts own parameters to achieve the optimal filtering effect, and once the statistical characteristics of the input signals change, the method can track the change, automatically adjusts the parameters, and enables the filter to reach the optimal again.
Signal x (k) =s (k) +v (k), where x (k) is the useful signal to be extracted and v (k) is the noise signal, the correlation with v (k) is not considered in the simulation. The wavelet is a Harr wavelet function, and multi-layer wavelet decomposition is performed according to analysis x (k) pairs. Signal signalThe noise of (2) is concentrated in a high frequency band, and the approximate component of the multi-layer filtering can remove part of the noise and is used as the input of the self-adaptive filter. The self-adaptive filtering algorithm adopts LMS algorithm, when the mean square error E [ E ] 2 (k)]The output x (k) will approach the signal s (k). And further, the error in the depression brain electrical signal can be reduced, so that the brain electrical signal is further extracted. Because the convergence rate of the LMS algorithm is sensitive to the distribution of the characteristic values of the autocorrelation function matrix of the signal, the strong correlation of the input signal can cause the dispersion of the characteristic values of the autocorrelation matrix of the input signal, namely, the maximum value and the minimum value of the characteristic values have larger difference, the convergence rate and the accuracy can be affected to a certain extent at the moment, an adaptive filtering algorithm based on wavelet decomposition and reconstruction are provided, the convergence rate of the adaptive filtering can be improved, the accuracy of acquiring the depression brain electrical signals is improved, and the noise of the signal is forcefully restrained.
Fig. 3 is a waveform of a signal denoised by wavelet combining adaptive filtering, where an original signal is decomposed into a plurality of subband signals of different scales by wavelet transformation, then each subband signal is subjected to adaptive filtering, and finally the processed signal is reconstructed into a denoised signal.
It is to be understood that the above-described embodiments of the present invention are merely illustrative of or explanation of the principles of the present invention and are in no way limiting of the invention. Accordingly, any modification, equivalent replacement, improvement, etc. made without departing from the spirit and scope of the present invention should be included in the scope of the present invention. Furthermore, the appended claims are intended to cover all such changes and modifications that fall within the scope and boundary of the appended claims, or equivalents of such scope and boundary.
Claims (5)
1. The method for extracting the depression brain electrical signal based on the wavelet and the adaptive filtering is characterized by comprising the following steps of:
step a, acquiring brain electrical signal data of a patient suffering from depression;
step b, preprocessing the signals by using a wavelet analysis method, selecting a proper wavelet basis function to carry out wavelet decomposition, filtering out useless physiological electric signals such as electrooculogram, myoelectricity, electrocardiograph and the like, and improving the signal-to-noise ratio of the signals;
step c, extracting the electroencephalogram signals by a self-adaptive filtering algorithm, and performing secondary wavelet transformation on the electroencephalogram signals;
step d, taking the reconstructed noise as a reference input of the adaptive filter, and performing adaptive filtering processing to obtain a denoising signal;
and e, reconstructing the signals obtained in the step d and the brain electrical signals of each frequency band in the step b to obtain the effective brain electrical signals of the patient suffering from depression.
2. The depression electroencephalogram signal extraction method based on wavelet and adaptive filtering according to claim 1, wherein the electroencephalogram signal data is collected, amplified and filtered by a signal collection unit, and then converted into a digital signal by an A/D converter to be sent to a real-time electroencephalogram signal processing module, and the real-time electroencephalogram signal processing module is used for eliminating noise caused by interference and physiological artifacts.
3. The method for extracting the depression electroencephalogram based on wavelet and adaptive filtering according to claim 2, wherein the step b is characterized in that a Mallat algorithm is adopted to realize wavelet decomposition, signals are decomposed into wavelet coefficients with different scales, and the basic formula of the Mallat algorithm is as follows:
where j is the scale of the wavelet transform, k is the index corresponding to the scale factor or wavelet coefficient, w j,k For the kth wavelet coefficient in the jth layer wavelet decomposition, n represents the position in the time domain, x j+1,2k-n For the 2k-n low frequency coefficients in the j+1 layer wavelet decomposition, y j+1,2k-n For the 2k-n high frequency coefficients in the j+1 layer wavelet decomposition, h (n) is the low pass filter coefficient at the n-th sampling point in the wavelet decomposition, g (n) is the high at the n-th sampling point in the wavelet decompositionAnd (5) passing filter coefficients.
4. A method for extracting an electroencephalogram signal from depression based on wavelet and adaptive filtering according to claim 3, wherein the method comprises the following steps: for the detail coefficient of each scale, low-pass filtering or high-pass filtering is realized by setting a threshold value so as to remove noise signals.
5. The method for extracting the depression electroencephalogram based on wavelet and adaptive filtering according to claim 4, wherein the method comprises the following steps of: wavelet reconstruction is carried out on the processed approximation coefficient and detail coefficient to obtain a denoised signal, and the basic formula of the wavelet reconstruction is as follows:
wherein x is n N is the position in time domain, J is the scale of wavelet transformation, k is the index corresponding to scale coefficient or wavelet coefficient, J is the highest number of layers of decomposition,for low pass filter coefficients at the j-th layer k instant,for the high-pass filter coefficient at the j-th layer k moment, a j,k,n For the approximation coefficient of the jth layer of the nth sampling point, d j,k,n Is the detail coefficient of the jth layer at the nth sampling point.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310660335.5A CN116671932A (en) | 2023-06-06 | 2023-06-06 | Depression brain electric signal extraction method based on wavelet and self-adaptive filtering |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310660335.5A CN116671932A (en) | 2023-06-06 | 2023-06-06 | Depression brain electric signal extraction method based on wavelet and self-adaptive filtering |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116671932A true CN116671932A (en) | 2023-09-01 |
Family
ID=87788549
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310660335.5A Pending CN116671932A (en) | 2023-06-06 | 2023-06-06 | Depression brain electric signal extraction method based on wavelet and self-adaptive filtering |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116671932A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117633442A (en) * | 2024-01-24 | 2024-03-01 | 小舟科技有限公司 | Brain-computer interface signal preprocessing method and computer equipment |
| CN117672457A (en) * | 2023-12-15 | 2024-03-08 | 山东奥洛瑞医疗科技有限公司 | Micro-feedback intelligent knee rehabilitation nursing method and device |
-
2023
- 2023-06-06 CN CN202310660335.5A patent/CN116671932A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117672457A (en) * | 2023-12-15 | 2024-03-08 | 山东奥洛瑞医疗科技有限公司 | Micro-feedback intelligent knee rehabilitation nursing method and device |
| CN117633442A (en) * | 2024-01-24 | 2024-03-01 | 小舟科技有限公司 | Brain-computer interface signal preprocessing method and computer equipment |
| CN117633442B (en) * | 2024-01-24 | 2024-05-10 | 小舟科技有限公司 | Brain-computer interface signal preprocessing method and computer equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Sun et al. | A novel end-to-end 1D-ResCNN model to remove artifact from EEG signals | |
| CN110353672B (en) | System and method for removing eye artifacts in electroencephalogram signals | |
| CN116671932A (en) | Depression brain electric signal extraction method based on wavelet and self-adaptive filtering | |
| CN106709469B (en) | Automatic sleep staging method based on electroencephalogram and myoelectricity multiple characteristics | |
| Nayak et al. | Filtering techniques for ECG signal processing | |
| CN108309290A (en) | The automatic removal method of Muscle artifacts in single channel EEG signals | |
| Li et al. | ECG denoising method based on an improved VMD algorithm | |
| CN108577834A (en) | A method of it is detected automatically for phase spike between epilepsy | |
| Patil et al. | Quality advancement of EEG by wavelet denoising for biomedical analysis | |
| CN109480832A (en) | The removing method of Muscle artifacts in a kind of single pass EEG signals | |
| AU2020103949A4 (en) | EEG Signal Mixed Noise Processing Method, Equipment and Storage Medium | |
| Bortolan et al. | Dynamic filtration of high-frequency noise in ECG signal | |
| Al-kadi et al. | Compatibility of mother wavelet functions with the electroencephalographic signal | |
| Wang et al. | Wavelet transform based ECG denoising using adaptive thresholding | |
| CN112426162A (en) | Fatigue detection method based on electroencephalogram signal rhythm entropy | |
| Sharma | EEG signal denoising based on wavelet transform | |
| CN112861625A (en) | Method for determining stacking denoising autoencoder model | |
| CN114403896A (en) | Method for removing ocular artifacts in single-channel electroencephalogram signal | |
| CN103816007A (en) | Equipment and method for tinnitus treatment based on electroencephalogram frequency-domain characteristic indexation algorithm | |
| Heydari et al. | Adaptive wavelet technique for EEG de-noising | |
| CN116172576A (en) | Electroencephalogram signal artifact removing method based on multi-module neural network | |
| CN114757236B (en) | Electroencephalogram signal denoising optimization method and system based on TQWT and SVMD | |
| CN111493864B (en) | EEG signal mixed noise processing method, equipment and storage medium | |
| Dembrani et al. | Accurate detection of ECG signals in ECG monitoring systems by eliminating the motion artifacts and improving the signal quality using SSG filter with DBE | |
| Murthy et al. | Design and implementation of hybrid techniques and DA-based reconfigurable FIR filter design for noise removal in EEG signals on FPGA |
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 |