CN108470182B - Brain-computer interface method for enhancing and identifying asymmetric electroencephalogram characteristics - Google Patents

Brain-computer interface method for enhancing and identifying asymmetric electroencephalogram characteristics Download PDF

Info

Publication number
CN108470182B
CN108470182B CN201810065848.0A CN201810065848A CN108470182B CN 108470182 B CN108470182 B CN 108470182B CN 201810065848 A CN201810065848 A CN 201810065848A CN 108470182 B CN108470182 B CN 108470182B
Authority
CN
China
Prior art keywords
electroencephalogram
brain
training set
computer interface
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.)
Active
Application number
CN201810065848.0A
Other languages
Chinese (zh)
Other versions
CN108470182A (en
Inventor
许敏鹏
肖晓琳
王仲朋
汤佳贝
陈龙
明东
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tianjin University
Original Assignee
Tianjin University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Tianjin University filed Critical Tianjin University
Priority to CN201810065848.0A priority Critical patent/CN108470182B/en
Publication of CN108470182A publication Critical patent/CN108470182A/en
Priority to US16/616,834 priority patent/US11221672B2/en
Priority to PCT/CN2018/125927 priority patent/WO2019144776A1/en
Application granted granted Critical
Publication of CN108470182B publication Critical patent/CN108470182B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • 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
    • 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
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235—Details of waveform analysis
    • A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/08—Feature extraction
    • 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

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Animal Behavior & Ethology (AREA)
  • Veterinary Medicine (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Public Health (AREA)
  • Signal Processing (AREA)
  • General Health & Medical Sciences (AREA)
  • Surgery (AREA)
  • Data Mining & Analysis (AREA)
  • Molecular Biology (AREA)
  • Psychiatry (AREA)
  • Biophysics (AREA)
  • Pathology (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Medical Informatics (AREA)
  • Physiology (AREA)
  • Mathematical Physics (AREA)
  • Evolutionary Biology (AREA)
  • Fuzzy Systems (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Psychology (AREA)
  • Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)

Abstract

The invention relates to a brain-computer interface method for enhancing and identifying asymmetric electroencephalogram characteristics, which comprises the following steps: step one, establishing a training set X through a brain-computer interface systemkTraining sample YlAnd an electroencephalogram signal module for testing the sample Y; secondly, performing frequency domain filtering and downsampling data processing on a test sample Y in the electroencephalogram signal module; thirdly, based on Fisher linear discrimination criterion, training set X in electroencephalogram signal modulekCalculating to obtain a spatial projection matrix W; step four, training set X in electroencephalogram signal modulekAnd the test sample Y is obtained by performing DSP spatial filtering according to the following formula
Figure DDA0001556649280000011
And WTA Y feature vector; step five, according to
Figure DDA0001556649280000012
And WTConstruction of projection matrix U by CCA spatial filtering of Y eigenvectorkAnd Vk(ii) a By obtaining feature vectors
Figure DDA0001556649280000013
WTY, projection matrix UkAnd VkCarrying out template matching according to the following formula to generate a feature vector rhol(ii) a Using different classifier models to pair feature vectors rholOutputting after identification; the method improves the signal-to-noise ratio of the electroencephalogram signal, thereby improving the classification and identification efficiency of the signal characteristics.

Description

Brain-computer interface method for enhancing and identifying asymmetric electroencephalogram characteristics
Technical Field
The invention relates to the technical field of brain-computer interface systems, in particular to a brain-computer interface method for enhancing and identifying asymmetric electroencephalogram characteristics.
Background
Brain-Computer Interface (BCI) is a system that directly converts central nervous system activities into artificial outputs, and can replace, repair, enhance, supplement, or improve the normal outputs of the central nervous system, thereby improving the interaction between the central nervous system and the internal and external environments. The method comprises the steps of collecting and analyzing electroencephalogram signals of a subject under different stimuli, and establishing an exchange and control channel between the human brain and a computer or other electronic equipment by using a certain engineering technical means. The BCI technology realizes a brand-new information interaction and control mode, and can provide a way for the disabled, particularly patients with impaired basic limb movement function but normal thinking, to communicate with the outside or operate the outside equipment without language or limb movement. For this reason, BCI technology is also gaining increasing attention. The P300-spinner based on the P300 characteristic in the Event-Related Potential (ERP) and the SSVEP-BCI based on the Steady-State Visual Evoked Potential (SSVEP) are brain-computer interface systems induced by Visual stimulation widely applied, and the Related technology thereof has been developed to be more stable and mature.
For real-time data acquisition systems, digital filtering is usually required to remove interference signals from acquired data, and conventional filtering methods usually filter out specific band frequencies, such as: low pass filtering, high pass filtering, band pass filtering, notch filtering, and the like. The electroencephalogram signal has the characteristics of nonlinearity and non-stationarity, in the research of a brain-computer interface system, how to process and analyze the acquired electroencephalogram signal, extracting weak electroencephalogram signal characteristics from complicated background electroencephalograms and classifying and identifying different characteristics are key factors for determining the performance of a BCI system, and because the electroencephalogram signal has frequency characteristics, a filtering means is also commonly used in the processing and analysis of the electroencephalogram signal, and the filtering frequency band can be adjusted according to different electroencephalogram characteristics. After filtering, methods for classifying and recognizing traditional brain electrical signals include Linear Discriminant Analysis (LDA), Common Spatial Pattern (CSP), Support Vector Machine (SVM), and typical Correlation Analysis (CCA). These methods all include the idea of spatial filtering, i.e. selecting one or several classification planes in a high-dimensional space, and using the vectors as spatial filters to spatially filter the signals, in order to reduce the high-dimensional signals to a low-dimensional one, which facilitates their classification. A typical correlation analysis algorithm is generally applied to an SSVEP-BCI system at present, and the algorithm is further improved by research, namely a self signal of a subject is introduced by applying a template matching principle in an electroencephalogram information processing process, so that the identification accuracy and the information transmission rate of the system are improved, and a powerful basis is laid for further converting a BCI technology to an application result.
Disclosure of Invention
The invention aims to overcome the defects in the background art and provide a brain-computer interface method for enhancing and identifying asymmetric electroencephalogram characteristics, which is a characteristic classification method combining a discrimination mode spatial filtering and a template matching principle, introduces a DSP spatial filtering method on the basis of the existing template matching CCA classification strategy, and constructs different decoding templates according to coding strategies of different stimulation paradigms so as to improve the signal-to-noise ratio of an electroencephalogram signal and improve the classification and identification efficiency of signal characteristics.
The invention is implemented by adopting the following technical scheme:
a brain-computer interface method for asymmetric electroencephalogram feature enhancement and identification comprises the following steps:
step one, establishing a training set X through a brain-computer interface systemkTraining sample YlAnd EEG signal data set of test sample Y
Secondly, performing frequency domain filtering and downsampling data processing on the electroencephalogram signal data centralized test sample Y;
thirdly, based on Fisher linear discrimination criterion, training set X in electroencephalogram signal modulekCalculating to obtain a spatial projection matrix W;
step four, carrying out centralized training set X on electroencephalogram signal datakAnd the test sample Y is obtained by performing DSP spatial filtering according to the following formula
Figure GDA0003423545520000021
And WTA Y feature vector;
Figure GDA0003423545520000022
SB＝∑11+∑22-∑12-∑21
Sw＝σ1 2+σ2 2
step five, according to
Figure GDA0003423545520000023
And WTThe Y eigenvector adopts the following formula to carry out CCA spatial filtering to construct a projection matrix UkAnd Vk；
Figure GDA0003423545520000024
Step six, obtaining the feature vector
Figure GDA0003423545520000025
WTY, projection matrix UkAnd VkCarrying out template matching according to the following formula to generate a feature vector rhol；
Figure GDA0003423545520000031
Step seven, adopting different classifier models to pair the eigenvector rholAnd outputting after identification.
The training set
Figure GDA0003423545520000032
k represents two types of features, namely k is 1, 2; the training sample
Figure GDA0003423545520000033
The test specimen
Figure GDA0003423545520000034
Where Nc represents the number of channels for acquiring brain waves, NtIndicating the length of the intercepted signal, NsRepresenting the number of training set samples.
Compared with the prior art, the invention has the advantages that:
1. the invention is used for the classification and identification of the asymmetric electroencephalogram characteristics, and can effectively improve the signal-to-noise ratio of the identification signal and improve the classification accuracy.
2. The method is used in the brain-computer interface system controlled by the asymmetric electroencephalogram characteristics, the average classification accuracy of the asymmetric electroencephalogram characteristics is improved by 17.88 percent compared with that of the traditional classification method, and the method proves that the brain-computer interface technology can be further perfected and the conversion of the technology to application results is promoted; the application range is wide.
3. The invention is applied to a brain-computer interface system based on asymmetric electroencephalogram characteristic control, and designs and implements BCI-spinner off-line and on-line brain-computer interface system experiments with an instruction set of 32; further research can obtain a perfect brain-computer interface system, and considerable social and economic benefits are expected to be obtained.
Drawings
FIG. 1 is a flow chart of a brain-computer interface method for asymmetric electroencephalogram feature enhancement and identification according to the present invention.
Fig. 2 is a schematic diagram of a brain-computer interface system including 32 instruction sets according to the present invention.
Detailed Description
The invention is further illustrated by the following specific examples and the accompanying drawings. The examples are intended to better enable those skilled in the art to better understand the present invention and are not intended to limit the present invention in any way.
As shown in FIG. 1, the present invention provides a brain-computer interface method for asymmetric electroencephalogram feature enhancement and identification;
step one 101, passing through the brain-machine interface system setup comprises training set XkTraining sample YlAnd a EEG data set of the test sample Y
Suppose that
Figure GDA0003423545520000041
For the training set, k represents two classes of features, i.e., k 1,2,
Figure GDA0003423545520000042
two types of training samples, 1,2,
Figure GDA0003423545520000043
to test a sample, wherein NcRepresenting the number of channels, N, through which the brain waves are acquiredtIndicating the length of the intercepted signal, NsRepresenting the number of training set samples. The training set and the test set are all processed with zero mean value on time scale, namely, the value s of each time pointtAll subtract the time window t1，t2]Time average of
Figure GDA0003423545520000044
As shown in equation (1):
Figure GDA0003423545520000045
averaging all samples in the training set to obtain a template signal of class k
Figure GDA0003423545520000046
And (4) showing. Covariance matrix between two types of templates
Figure GDA0003423545520000047
Expressed as:
Figure GDA0003423545520000048
two kinds of signals X1And X2The variances of (a) are expressed as:
Figure GDA0003423545520000049
Figure GDA00034235455200000410
step two 102, selecting electroencephalogram signal data sets
Figure GDA00034235455200000411
Carrying out frequency domain filtering and down-sampling data processing on the test sample;
step three 103, based on Fisher linear discrimination criterion, training set X in electroencephalogram signal modulekCalculating to obtain a spatial projection matrix W;
step four 104, carrying out centralized training set X on the electroencephalogram signal datakAnd the test sample Y is obtained by performing DSP spatial filtering according to the following formula
Figure GDA00034235455200000412
And WTY feature vector
Based on Fisher linear discriminant criterion, a DSP algorithm obtains a projection matrix W to enable two types of characteristic signals to have larger separability after projection, the matrix W can be used as a spatial filter, and the solving method is as follows:
Figure GDA00034235455200000413
SB＝∑11+∑22-∑12-∑21 (6)
Sw＝σ1 2+σ2 2 (7)
wherein λiIs the eigenvector of the ith column in the matrix W, NWIndicating the number of spatial filters selected. Common-mode signals between the two types of signals can be filtered out through W spatial filtering, and the application of the CCA algorithm can be realized through constructing two projectionsMatrix UkAnd VkTo calculate after DSP spatial filtering
Figure GDA0003423545520000051
And WTCorrelation between Y, CCA spatial filter UkAnd VkCalculated from the following formula (8).
Step five 105, according to
Figure GDA0003423545520000052
And WTThe Y eigenvector adopts the following formula to carry out CCA spatial filtering to construct a projection matrix UkAnd Vk
Figure GDA0003423545520000053
Wherein,
Figure GDA0003423545520000054
representing a mathematical expectation. Typical correlation analysis is a statistical analysis method that measures a linear correlation between two multidimensional variables. Unlike fitting a sample point with a straight line in linear regression, CCA regards multidimensional feature vectors as a whole, and finds a set of optimal solutions by using a mathematical method, so that the two whole have the maximum associated weight, i.e., the value calculated by equation (8) is the maximum, which is the purpose of typical correlation analysis.
Step six 106, obtaining the feature vector
Figure GDA0003423545520000055
WTY, projection matrixUk and VkCarrying out template matching according to the following formula to generate a feature vector rhol；
In the template matching process, a template is constructed by training set data, and the template construction can be correspondingly adjusted according to different stimulation modes, wherein the vector rho shown in a formula (9) is taken as an example for classifying the asymmetric electroencephalogram characteristic signalskRepresenting the similarity between the training template and the training sample signal i.
Figure GDA0003423545520000056
Where corr (×) represents the pearson correlation coefficient and dist (×) represents the euclidean distance. If ρk1，ρk2，ρk3，ρk4And ρk5The larger the value, the more YlAnd
Figure GDA0003423545520000057
the greater the correlation between them. Connection rho1·，lAnd ρ2·，lObtaining a characteristic vector rho obtained by extracting the characteristics of the training template and the training sampleMExpressed by equation (10):
ρl＝(ρ1·l，ρ2.l)，l＝1，2 (10)
step seven 107, adopting different classifier models to pair the eigenvector rholAnd outputting after identification.
From the eigenvectors ρlDifferent classifier models of different mode recognition algorithms such as Linear Discriminant Analysis (LDA) and Support Vector Machine (SVM) are established, a test sample Y is sent to a classifier for mode recognition after being preprocessed and subjected to feature extraction, and then the class of the sample is predicted and a result is output, as shown in figure 1.
Fig. 2 is a schematic structural diagram of a brain-computer interface system including 32 instruction sets for the algorithm application of the present invention. The system comprises an electroencephalogram acquisition system such as a liquid crystal display stimulation interface, an electroencephalogram electrode, an electroencephalogram amplifier and the like, a computer processing platform and the like. The system induces two types of asymmetric electroencephalogram characteristics by applying a visual stimulation paradigm, acquires electroencephalogram signals by adopting an electroencephalogram digital acquisition system produced by Neuroscan company, amplifies and filters the signals by an electroencephalogram amplifier and then inputs the signals into a computer, classifies the two types of electroencephalogram characteristics by applying the algorithm, and finally decodes the electroencephalogram signals and converts the signals into BCI instructions to output. And the stimulation presentation and the data processing analysis are completed based on the Matlab platform.
Calculating the signal-to-noise ratios (SNR) of two types of asymmetric electroencephalogram characteristic signals to be-17.98 dB and-14.90 dB respectively, wherein the SNR is defined as the ratio of signal energy to noise energy, and the calculation formula is as follows:
Figure GDA0003423545520000061
wherein AMP isiRepresents the average amplitude of the signal within the ith trial time window and N represents the number of trials.
The BCI system applied by the algorithm of the present invention was tested on 12 subjects. The experimental result shows that after the algorithm is applied, the average classification accuracy of 12 tested subjects is improved by 17.88 percent, and the significance is improved (the paired T test result is T: T)11P is less than 0.01) and the signal-to-noise ratio of the two types of characteristic signals is respectively improved to-9.71 dB and-8.68 dB after being filtered by DSP space.
It should be understood that the embodiments and examples discussed herein are illustrative only and that various modifications or changes in light thereof will be suggested to persons skilled in the art and are to be included within the spirit and purview of this application and scope of the appended claims.

Claims (2)

1. A brain-computer interface method for asymmetric electroencephalogram feature enhancement and identification is characterized by comprising the following steps:
step one, establishing a training set X through a brain-computer interface systemkTraining sample YlAnd an electroencephalogram signal dataset of the test sample Y;
secondly, performing frequency domain filtering and downsampling data processing on the electroencephalogram signal data centralized test sample Y;
thirdly, based on Fisher linear discrimination criterion, training set X in electroencephalogram signal modulekCalculating to obtain a spatial projection matrix W;
step four, carrying out centralized training set X on electroencephalogram signal datakAnd the test sample Y is obtained by performing DSP spatial filtering according to the following formula
Figure FDA0003488644760000011
And WTA Y feature vector;
Figure FDA0003488644760000012
SB＝∑11+∑22-∑12-∑21
Sw＝σ1 2+σ2 2
step five, according to
Figure FDA0003488644760000013
And WTThe Y eigenvector adopts the following formula to carry out CCA spatial filtering to construct a projection matrix UkAnd Vk；
Figure FDA0003488644760000014
Step six, obtaining the feature vector
Figure FDA0003488644760000015
WTY, projection matrix UkAnd VkCarrying out template matching according to the following formula to generate a feature vector rhol；
Figure FDA0003488644760000016
Wherein: corr (, p) denotes the pearson correlation coefficient, dist (, p) denotes the euclidean distance;
step seven, adopting different classifier models to pair the eigenvector rholAnd outputting after identification.
2. The brain-computer interface method for asymmetric electroencephalogram feature enhancement and identification according to claim 1The method is characterized in that: the training set
Figure FDA0003488644760000017
k represents two types of features, namely k is 1, 2; the training sample
Figure FDA0003488644760000018
The test specimen
Figure FDA0003488644760000019
Wherein N iscRepresenting the number of channels, N, through which the brain waves are acquiredtIndicating the length of the intercepted signal, NsRepresenting the number of training set samples.
CN201810065848.0A 2018-01-23 2018-01-23 Brain-computer interface method for enhancing and identifying asymmetric electroencephalogram characteristics Active CN108470182B (en)

Priority Applications (3)

Application Number Priority Date Filing Date Title
CN201810065848.0A CN108470182B (en) 2018-01-23 2018-01-23 Brain-computer interface method for enhancing and identifying asymmetric electroencephalogram characteristics
US16/616,834 US11221672B2 (en) 2018-01-23 2018-12-30 Asymmetric EEG-based coding and decoding method for brain-computer interfaces
PCT/CN2018/125927 WO2019144776A1 (en) 2018-01-23 2018-12-30 Coding-decoding method for brain-machine interface system based on asymmetric electroencephalographic features

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810065848.0A CN108470182B (en) 2018-01-23 2018-01-23 Brain-computer interface method for enhancing and identifying asymmetric electroencephalogram characteristics

Publications (2)

Publication Number Publication Date
CN108470182A CN108470182A (en) 2018-08-31
CN108470182B true CN108470182B (en) 2022-04-29

Family

ID=63266061

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810065848.0A Active CN108470182B (en) 2018-01-23 2018-01-23 Brain-computer interface method for enhancing and identifying asymmetric electroencephalogram characteristics

Country Status (1)

Country Link
CN (1) CN108470182B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2019144776A1 (en) * 2018-01-23 2019-08-01 天津大学 Coding-decoding method for brain-machine interface system based on asymmetric electroencephalographic features
CN111158462B (en) * 2019-11-28 2021-02-02 燕山大学 Method for improving electroencephalogram wakefulness based on implementation boundary evasion task model
CN113591598B (en) * 2021-07-07 2024-07-12 河北工业大学 Correlation analysis-based brain-computer interface cross-load linear discrimination method
CN113705732B (en) * 2021-09-26 2024-02-02 华东理工大学 Method and device for reducing P300 training time based on general model

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1776572A (en) * 2005-12-08 2006-05-24 清华大学 Computer man-machine interacting method based on steady-state vision induced brain wave
CN101049236A (en) * 2007-05-09 2007-10-10 西安电子科技大学 Instant detection system and detection method for state of attention based on interaction between brain and computer
CN101159086A (en) * 2007-11-22 2008-04-09 中国人民解放军国防科学技术大学 Calling device based on brain electric information demodulation
CN106502410A (en) * 2016-10-27 2017-03-15 天津大学 Improve the transcranial electrical stimulation device of Mental imagery ability and method in brain-computer interface

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11269891B2 (en) * 2014-08-21 2022-03-08 Affectomatics Ltd. Crowd-based scores for experiences from measurements of affective response

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1776572A (en) * 2005-12-08 2006-05-24 清华大学 Computer man-machine interacting method based on steady-state vision induced brain wave
CN101049236A (en) * 2007-05-09 2007-10-10 西安电子科技大学 Instant detection system and detection method for state of attention based on interaction between brain and computer
CN101159086A (en) * 2007-11-22 2008-04-09 中国人民解放军国防科学技术大学 Calling device based on brain electric information demodulation
CN106502410A (en) * 2016-10-27 2017-03-15 天津大学 Improve the transcranial electrical stimulation device of Mental imagery ability and method in brain-computer interface

Also Published As

Publication number Publication date
CN108470182A (en) 2018-08-31

Similar Documents

Publication Publication Date Title
Li et al. Depression recognition using machine learning methods with different feature generation strategies
Aslan et al. Automatic Detection of Schizophrenia by Applying Deep Learning over Spectrogram Images of EEG Signals.
CN108470182B (en) Brain-computer interface method for enhancing and identifying asymmetric electroencephalogram characteristics
CN107844755B (en) Electroencephalogram characteristic extraction and classification method combining DAE and CNN
WO2018014436A1 (en) Emotion eeg recognition method providing emotion recognition model time robustness
US11221672B2 (en) Asymmetric EEG-based coding and decoding method for brain-computer interfaces
CN110353673B (en) Electroencephalogram channel selection method based on standard mutual information
CN114533086B (en) Motor imagery brain electrolysis code method based on airspace characteristic time-frequency transformation
Lopes et al. Automatic electroencephalogram artifact removal using deep convolutional neural networks
CN109375776B (en) Electroencephalogram action intention recognition method based on multi-task RNN model
CN101828921A (en) Identity identification method based on visual evoked potential (VEP)
CN114578963B (en) Electroencephalogram identity recognition method based on feature visualization and multi-mode fusion
CN109222966A (en) A kind of EEG signals sensibility classification method based on variation self-encoding encoder
Gao et al. EEG classification based on sparse representation and deep learning
CN112861629B (en) Multi-window distinguishing typical pattern matching method and brain-computer interface application
CN114330422A (en) Cross-test migration learning method for estimating electroencephalogram emotional characteristics in real time
CN115414050A (en) EEG brain network maximum clique detection method and system for realizing emotion recognition
Li et al. Feature extraction based on high order statistics measures and entropy for eeg biometrics
Prabhakar et al. EM based non-linear regression and singular value decomposition for epilepsy classification
CN111616680A (en) Automatic mental load identification method and system
Zhou et al. A smart universal single-channel blind source separation method and applications
CN113907770B (en) Ratchet composite wave detection and identification method and system based on feature fusion
CN109117790B (en) Brain print identification method based on frequency space index
CN114358090B (en) Motor imagery electroencephalogram signal classification method based on PSD and CSP
CN113017648B (en) Electroencephalogram signal identification method and system

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