CN112494043A - Electroencephalogram signal detection method based on tensor method - Google Patents
Electroencephalogram signal detection method based on tensor method Download PDFInfo
- Publication number
- CN112494043A CN112494043A CN202011261249.XA CN202011261249A CN112494043A CN 112494043 A CN112494043 A CN 112494043A CN 202011261249 A CN202011261249 A CN 202011261249A CN 112494043 A CN112494043 A CN 112494043A
- Authority
- CN
- China
- Prior art keywords
- psychological
- matrix
- electroencephalogram
- parameter
- electroencephalogram signal
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 19
- 238000001514 detection method Methods 0.000 title claims abstract description 9
- 239000011159 matrix material Substances 0.000 claims abstract description 58
- 238000004458 analytical method Methods 0.000 claims abstract description 12
- 230000002159 abnormal effect Effects 0.000 claims abstract description 10
- 238000003745 diagnosis Methods 0.000 claims abstract description 6
- 238000011160 research Methods 0.000 abstract description 6
- 210000004556 brain Anatomy 0.000 description 7
- 230000001107 psychogenic effect Effects 0.000 description 5
- 238000000605 extraction Methods 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 230000005856 abnormality Effects 0.000 description 3
- 206010012374 Depressed mood Diseases 0.000 description 2
- 238000000354 decomposition reaction Methods 0.000 description 2
- 230000036651 mood Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000000506 psychotropic effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 230000008451 emotion Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 208000020016 psychiatric disease Diseases 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000017105 transposition Effects 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/16—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
- A61B5/72—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
- A61B5/7257—Details of waveform analysis characterised by using transforms using Fourier transforms
Abstract
The invention discloses an electroencephalogram signal detection method based on a tensor method, which comprises the following steps: step1, establishing a PARAFAC model based on the acquired electroencephalogram signals; step2, extracting the psychological parameter characteristics of the electroencephalogram signals by adopting an alternating least square method to obtain a psychological parameter matrix, an electroencephalogram tester channel characteristic matrix and an acquisition environment matrix; step3, performing identifiability condition analysis on the feature of the psychological parameter through the uniqueness of the PARAFAC model; and 4, performing fast Fourier transform on a psychological generation mechanism and a psychological abnormal factor parameter of the electroencephalogram signal psychological parameter matrix to obtain an amplitude-frequency characteristic curve of the psychological parameter for later psychological diagnosis and treatment. The tensor has a research space in the expression and analysis of high-dimensional data, and the internal structure of some data cannot be damaged; the psychological characteristic parameters of the electroencephalogram signals are accurately detected and extracted.
Description
Technical Field
The invention belongs to the technical field of electroencephalogram signal detection, and particularly relates to an electroencephalogram signal detection method based on a tensor method.
Background
With the development of information science and technology, the idea that a skyscraper runs empty in the brain of some people is slowly becoming a reality. Among them, electroencephalogram has been gradually applied to real life, and electroencephalogram is a signal generated by brain neuron activity, and contains abundant brain state information and a large amount of physiological information related to human body, and electroencephalogram information is receiving more and more attention from people. Research shows that the electroencephalogram signals with different psychological changes are different, and when the mind or emotion changes, the electroencephalogram signal components related to the psychology also change, and the psychological changes of patients with psychological disorders can be observed by reading the electroencephalogram signals.
In the traditional method, data is generally expressed in a matrix form, and even if the original data is in a high-order structure, the original data must be expanded into the matrix form for processing.
Disclosure of Invention
The invention aims to provide an electroencephalogram signal detection method based on a tensor method, which aims to solve the problems that the structure of original data can be damaged, the internal correlation of the data can be damaged and the like in the background technology.
The invention provides an electroencephalogram signal detection method based on a tensor method, which comprises the following steps:
step1, establishing a PARAFAC model based on the acquired electroencephalogram signals;
step2, extracting the psychological parameter characteristics of the electroencephalogram signals by adopting an alternating least square method to obtain a psychological parameter matrix, an electroencephalogram tester channel characteristic matrix and an acquisition environment matrix;
step3, performing identifiability condition analysis on the feature of the psychological parameter through the uniqueness of the PARAFAC model;
and 4, performing fast Fourier transform on a psychological generation mechanism and a psychological abnormal factor parameter of the electroencephalogram signal psychological parameter matrix to obtain an amplitude-frequency characteristic curve of the psychological parameter for later psychological diagnosis and treatment.
Compared with the prior art, the invention has the beneficial effects that:
the method establishes a PARAFAC model containing a channel characteristic matrix, a psychological characteristic parameter matrix and an acquisition environment matrix for the electroencephalogram signals. The extraction of the psychological characteristic parameters of the electroencephalogram signals is realized by using the PARAFAC model uniqueness theorem. FFT conversion is carried out on the specific channel to obtain an amplitude-frequency characteristic curve graph of each channel, and the mutual influence among the parameters can be analyzed according to the amplitude-frequency characteristic curve graph of each channel. Compared with other inventions, the method has the following obvious advantages: (1) tensors have research space in the expression and analysis of high-dimensional data, and the internal structure of some data cannot be damaged; (2) the psychological characteristic parameters of the electroencephalogram signals are accurately detected and extracted.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a graph of SNR-NMSE of a channel characterization parameter of the present invention;
FIG. 3 is a graph of the amplitude-frequency characteristic of the psychogenic mechanism of the present invention;
FIG. 4 is a graph showing the magnitude-frequency characteristics of the psychotropic factors of the present invention;
FIG. 5 is a graph of magnitude-frequency characteristics of the psychotropic factors of the present invention.
Detailed Description
The present invention is described in detail with reference to the embodiments shown in the drawings, but it should be understood that these embodiments are not intended to limit the present invention, and those skilled in the art should understand that functional, methodological, or structural equivalents or substitutions made by these embodiments are within the scope of the present invention.
The embodiment provides an electroencephalogram signal detection method based on a tensor method, which comprises the following steps:
step1, establishing a PARAFAC model based on the acquired electroencephalogram signals;
step2, performing psychological parameter feature extraction on the electroencephalogram signal by adopting an alternating least square method (ALS) to obtain a psychological parameter matrix, an electroencephalogram tester channel characteristic matrix and an acquisition environment matrix;
step3, performing identifiability condition analysis on the feature of the psychological parameter through the uniqueness of the PARAFAC model;
and 4, performing fast Fourier transform on a psychological generation mechanism and a psychological abnormal factor parameter of the electroencephalogram signal psychological parameter matrix to obtain an amplitude-frequency characteristic curve of the psychological parameter for later psychological diagnosis and treatment.
The invention has obvious advantages compared with other inventions, including:
(1) the tensor has a research space in the expression and analysis of high-dimensional data, the internal structure of some data cannot be damaged, and the tensor can model the data on the premise of not damaging the internal structure of some data, so that the tensor has a certain research space in the expression and analysis of a multi-dimensional state;
(2) the method can accurately detect and extract the psychological characteristic parameters of the electroencephalogram signals, is used as a multi-modal analysis tool, and tensor decomposition is also started to be used for electroencephalogram signal analysis, so that the method becomes a new research direction after time domain, frequency domain, time frequency and space-time analysis.
To verify the performance of the proposed algorithm, the method was analyzed using a MATLAB simulation program. It should be noted that: once an item is defined in a drawing, it need not be further defined and explained in subsequent drawings.
Referring to fig. 1, fig. 1 is a flow chart of the present invention. The example is particularly applicable to a multi-dimensional state study space.
Step1, assuming that the channel characteristic of the electroencephalogram tester is represented by a, the psychological parameter of the tested human electroencephalogram signal is represented by b, and the electroencephalogram signal acquisition environment is represented by c, the acquired electroencephalogram signal can be written as follows:
wherein the content of the first and second substances,
noise and interference in the received electroencephalogram signals; i.e. i1、i2Respectively representing the number of channels at the receiving end of the sampling test and the corresponding ith2An individual subject; r is1、r2Respectively represent the r-th1The number of channels at the sending end of the test and the number of the psychological abnormal factors. The above formula can alsoWritten in the following form:
wherein i2＝1,2,…,I2，
Channel characteristic matrix of brain electrical tester
Brain electrical signal psychological parameter matrix
Collection environment matrix
The elements (A) and (B) in (B),
for the purpose of matrix diagonalization operators,
for the ith in the received brain electrical signal2Noise matrix of individual components.
Step2, assume that the collected EEG signal is
Firstly, a PARAFAC model is established for the collected electroencephalogram signal X, and the electroencephalogram signal for establishing the PARAFAC model is subjected to low-rank decomposition by utilizing an ALS algorithm. Wherein, the section operation is carried out on the EEG signal X measured by the EEG tester along the X direction to obtain the ith sub-section
Wherein, (.)TDenotes transposition, diag (. circle.) as diagonalization operator, diag (A)i) Represents a diagonal matrix formed by the ith row elements of the electroencephalogram signal parameter matrix A measured by the electroencephalogram tester. Arranging all the sub-section matrixes in rows to obtain a section matrix along the x direction
Wherein [ ] indicates a Khatri-Rao product. Similarly, according to the formula (3), the electroencephalogram signal X measured by the electroencephalogram tester is respectively subjected to section interception along the y direction and the z direction to obtain a section matrix
And
X2＝(B⊙C)AT (5)
X3＝(C⊙A)BT (6)
and for the EEG signal X after the section1、X2、X3Performing ALS algorithm iterative estimation processing to obtain a channel characteristic matrix of the electroencephalogram tester
Matrix of tested human brain electrical signal
And collecting the environment matrix
And finally, carrying out FFT (fast Fourier transform) on the psychology generation mechanism of the extracted electroencephalogram signal and the psychology abnormal factor psychology parameter channel to obtain an amplitude-frequency characteristic curve of the psychology parameters, and preparing for psychology diagnosis and treatment.
Step3, establishing a PARAFAC model for the electroencephalogram signals, wherein the channel characteristic matrix A, the tested human electroencephalogram signal matrix B and the acquisition environment matrix C of the electroencephalogram tester are respectively as follows:
C＝[(B⊙A)-1·X3]T (7)
A＝[(C⊙B)-1·X2]T (8)
B＝[(A⊙C)-1·X1]T (9)
step4, identifiability Condition analysis, representing k-rank of matrix A as kAEven if any k of AAColumn-linearly independent maximum integer kA. Similarly, k-rank of matrix B is denoted as kBK-rank of the matrix C is denoted as kCAvailable according to the PARAFAC model uniqueness theorem.
kA+kB+kC≥2(R+1) (15)
Wherein R is the dimension of the electroencephalogram signal column. Satisfying the above formula A, B, C decomposes uniquely.
The Step5 and ALS algorithms can respectively solve a channel characteristic matrix A, an electroencephalogram signal matrix B and an acquisition environment matrix C of the electroencephalogram tester. The method comprises the following implementation steps:
(1) initializing a channel characteristic matrix A, an electroencephalogram signal matrix B and an acquisition environment matrix C of the electroencephalogram tester.
(2) Calculating the starting errors fit and fitold, wherein
fitold＝2*fit (11)
(3) Firstly, according to a formula (3), an initialized electroencephalogram signal matrix B and a channel characteristic matrix A of an electroencephalogram tester are applied to estimate an acquisition environment matrix of the electroencephalogram signals to obtain:
secondly, according to the formula (4), the acquisition matrix estimation value of the obtained electroencephalogram signal is utilized
And an initialized electroencephalogram signal matrix B, estimating a channel characteristic matrix of the electroencephalogram tester to obtain:
thirdly, according to the formula (5), the estimated value of the acquisition environment matrix of the obtained electroencephalogram signal is applied
Estimation of channel characteristic matrix of electroencephalographic tester
Estimating the electroencephalogram signal matrix B to obtain:
(4) the errors fit and fitold are recalculated.
(5) Repeating the steps (3) to (4) until the condition of | (fit-fitold)/fitold | > SMALLNUMBER is satisfied, wherein SMALLNUMBER ═ eps.
(6) And carrying out matrix blurring and column blurring elimination operations on the obtained result.
(7) And extracting the electroencephalogram signal matrix B. The SNR-NMSE curve of the electroencephalogram tester channel characteristic parameter obtained by the ALS algorithm, as shown in fig. 2, can see that the normalized mean square error of the channel characteristic parameter decreases with the increase of the signal-to-noise ratio, which indicates that the extraction of the electroencephalogram tester channel characteristic matrix is correct.
(8) And performing FFT (fast Fourier transform) on the psychological generation mechanism and the psychological abnormal factor parameters of the tested human brain electrical signals to obtain an amplitude-frequency characteristic curve of the psychological parameters so as to facilitate the psychological diagnosis and treatment in the later period.
Psychological generation mechanism and psychological abnormal factor e of electroencephalogram signal of tested person1、e2The amplitude-frequency characteristic curves are shown in FIGS. 3-5. The psychological generation mechanism of the electroencephalogram signals is obviously changed between 0Hz and 20Hz, and the tested person has a mood of low mood during the test. Accompanied by the change of the psychogenic mechanism, a psychogenic abnormality factor e1And e2Corresponding changes have also occurred. Wherein the psychological abnormality factor e1Large variation range and psychological abnormal factor e2The variation amplitude is small. The change of the psychogenic mechanism in the frequency band is related to the psychogenic abnormality factor e1The influence is large, and besides large amplitude change between 0Hz and 40Hz, small amplitude fluctuation also occurs between 40Hz and 120 Hz. The person to be tested experiences a lag in mood depression and a gradual tendency towards calm mind during the test, with small amplitude alternating changes in depression, calm and excited mood. Therefore, the psychological generation mechanism and the parameter of the psychological abnormal factor fluctuate correspondingly. Because the measurable frequency range of the electroencephalogram signals is about 0-160 Hz, the amplitude of the amplitude-frequency characteristic curve graph is close to 0 after 160 Hz. Simulation results show that the channel characteristic parameters are normalized. The normalized mean square error is reduced along with the increase of the signal-to-noise ratio, and the correctness of the extraction of the electroencephalogram signal tester channel characteristic matrix is proved.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.
Claims (1)
1. An electroencephalogram signal detection method based on a tensor method is characterized by comprising the following steps:
step1, establishing a PARAFAC model based on the acquired electroencephalogram signals;
step2, extracting the psychological parameter characteristics of the electroencephalogram signals by adopting an alternating least square method to obtain a psychological parameter matrix, an electroencephalogram tester channel characteristic matrix and an acquisition environment matrix;
step3, performing identifiability condition analysis on the feature of the psychological parameter through the uniqueness of the PARAFAC model;
and 4, performing fast Fourier transform on a psychological generation mechanism and a psychological abnormal factor parameter of the electroencephalogram signal psychological parameter matrix to obtain an amplitude-frequency characteristic curve of the psychological parameter for later psychological diagnosis and treatment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011261249.XA CN112494043B (en) | 2020-11-12 | 2020-11-12 | Electroencephalogram signal detection method based on tensor method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011261249.XA CN112494043B (en) | 2020-11-12 | 2020-11-12 | Electroencephalogram signal detection method based on tensor method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112494043A true CN112494043A (en) | 2021-03-16 |
| CN112494043B CN112494043B (en) | 2023-01-06 |
Family
ID=74957234
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011261249.XA Active CN112494043B (en) | 2020-11-12 | 2020-11-12 | Electroencephalogram signal detection method based on tensor method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112494043B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101947356A (en) * | 2010-10-22 | 2011-01-19 | 上海交通大学 | Injured brain function rehabilitation device based on brain-computer interaction |
| US20140236039A1 (en) * | 2011-10-21 | 2014-08-21 | Commissariat A I'energie Atomique Et Aux Energies Alternatives | Method of Calibrating and Operating a Direct Neural Interface System |
| CN105160154A (en) * | 2015-08-07 | 2015-12-16 | 武汉大学 | Parallel factor based multidimensional data analysis method |
| CN107645460A (en) * | 2017-08-25 | 2018-01-30 | 长江大学 | The multipath parameter evaluation method that real value parallel factor decomposes |
| CN110013248A (en) * | 2018-01-08 | 2019-07-16 | 上海交通大学 | Brain electricity tensor mode identification technology and brain-machine interaction rehabilitation system |
-
2020
- 2020-11-12 CN CN202011261249.XA patent/CN112494043B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101947356A (en) * | 2010-10-22 | 2011-01-19 | 上海交通大学 | Injured brain function rehabilitation device based on brain-computer interaction |
| US20140236039A1 (en) * | 2011-10-21 | 2014-08-21 | Commissariat A I'energie Atomique Et Aux Energies Alternatives | Method of Calibrating and Operating a Direct Neural Interface System |
| CN105160154A (en) * | 2015-08-07 | 2015-12-16 | 武汉大学 | Parallel factor based multidimensional data analysis method |
| CN107645460A (en) * | 2017-08-25 | 2018-01-30 | 长江大学 | The multipath parameter evaluation method that real value parallel factor decomposes |
| CN110013248A (en) * | 2018-01-08 | 2019-07-16 | 上海交通大学 | Brain electricity tensor mode identification technology and brain-machine interaction rehabilitation system |
Non-Patent Citations (1)
| Title |
|---|
| 刘华生等: "基于张量分解的运动想象脑电分类算法", 《微计算机信息》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112494043B (en) | 2023-01-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Sangaiah et al. | An intelligent learning approach for improving ECG signal classification and arrhythmia analysis | |
| WO2019038109A1 (en) | A method of detecting abnormalities in ecg signals | |
| WO2007130997A2 (en) | Method and device for filtering, segmenting, compressing and classifying oscillatory signals | |
| Veer | A technique for classification and decomposition of muscle signal for control of myoelectric prostheses based on wavelet statistical classifier | |
| KR101366101B1 (en) | System and method for classificating normal signal of personalized ecg | |
| Kang et al. | A method of denoising multi-channel EEG signals fast based on PCA and DEBSS algorithm | |
| US20140025715A1 (en) | Neural Signal Processing and/or Interface Methods, Architectures, Apparatuses, and Devices | |
| US11638562B2 (en) | Brain activity analysis method and apparatus thereof | |
| Jaswal et al. | QRS detection using wavelet transform | |
| CN107595247A (en) | A kind of monitoring method and system of the depth of anesthesia based on EEG signals | |
| CN109044348A (en) | atrial fibrillation detection device and storage medium | |
| Tang et al. | ECG de-noising based on empirical mode decomposition | |
| Zhang et al. | Non-invasive detection of low-level muscle fatigue using surface EMG with wavelet decomposition | |
| CN116361688A (en) | Multi-mode feature fusion model construction method for automatic classification of electrocardiographic rhythms | |
| CN115486849A (en) | Electrocardiosignal quality evaluation method, device and equipment | |
| CN112336369B (en) | Coronary heart disease risk index evaluation system of multichannel heart sound signals | |
| CN112494043B (en) | Electroencephalogram signal detection method based on tensor method | |
| CN109002798B (en) | Single-lead visual evoked potential extraction method based on convolutional neural network | |
| Hegde et al. | A review on ECG signal processing and HRV analysis | |
| CN115553784A (en) | Coronary heart disease assessment method and system based on electrocardio and heart sound signal coupling analysis | |
| CN115067878A (en) | EEGNet-based resting state electroencephalogram consciousness disorder classification method and system | |
| Jokić et al. | An efficient ECG modeling for heartbeat classification | |
| Khandait et al. | ECG signal processing using classifier to analyses cardiovascular disease | |
| Prashar et al. | Morphology analysis and time interval measurements using mallat tree decomposition for CVD Detection | |
| Das et al. | On an algorithm for detection of QRS complexes in noisy electrocardiogram signal |
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 |