CN108875580A - A kind of multiclass Mental imagery EEG signal identification method based on because imitating network - Google Patents
A kind of multiclass Mental imagery EEG signal identification method based on because imitating network Download PDFInfo
- Publication number
- CN108875580A CN108875580A CN201810463020.0A CN201810463020A CN108875580A CN 108875580 A CN108875580 A CN 108875580A CN 201810463020 A CN201810463020 A CN 201810463020A CN 108875580 A CN108875580 A CN 108875580A
- Authority
- CN
- China
- Prior art keywords
- network
- mental imagery
- imitating
- multivariable
- matrix
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
-
- 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
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- 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
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Signal Processing (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The invention discloses a kind of multiclass Mental imagery EEG signal identification methods based on because imitating network；Specially 1, it is acquired using brain wave acquisition equipment and drives EEG signals；2, collected EEG signals are pre-processed, including frequency reducing, noise reduction；3, building Mental imagery is because imitating network；4, to Mental imagery because effect network matrix is encoded；5, classified to Mental imagery because imitating the feature of network using linear SVM；The present invention need to only use Mental imagery because of an effect network matrix convolutional neural networks of training, and using 8 characteristic value training once linear support vector machine classifiers, we can be carried out online recognition multiclass Mental imagery EEG signals.
Description
Technical field
The invention belongs to bioelectrical signals to identify field, be related to multiclass Mental imagery EEG signal identification method, specifically
A kind of multiclass Mental imagery EEG signal identification method based on because imitating network.
Background technique
Mental imagery is a kind of special motor function state, can be expressed as people be carried out some movement carried out it is interior
The heart is rehearsed, and due to not causing the activity of limbs, muscle, there is apparent differences for it and actual motion.Mental imagery conduct
Most one of the rehabilitation new method of potential applicability in clinical practice in recent years, causes the concern of more and more researchers, has
Studies have shown that it can effectively activation moves relevant cerebral cortex region as actual motion.
With the continuous development and progress of artificial intelligence technology, healing robot gradually becomes clinical health in the world
One of important technical treated again.In conjunction with Mental imagery healing robot can assist cerebral injury quadriplegia patient into
Row rehabilitation training, this technology bring Gospel to cerebral injury quadriplegia patient.But the movement of online recognition patient at present
The identification of imagination task and multiclass Mental imagery EEG signals has become urgent problem to be solved in healing robot design,
It needs further to study and the expansion in method.
Summary of the invention
In view of the above problems, proposing a kind of multiclass Mental imagery EEG signal identification method based on because imitating network matrix.
This method has the characteristics that off-line training, online recognition.
According to technical solution provided by the invention, a kind of multiclass Mental imagery EEG signals based on because imitating network are proposed
Recognition methods, including following 5 steps：
Step 1, the multiclass Mental imagery EEG signals that 64 leads are acquired using brain wave acquisition equipment；
Step 2 pre-processes collected EEG signals, including frequency reducing, noise reduction.
Two frequency bands in each channel are extracted, are α (8-15HZ), β (16-30HZ) respectively.Because during Mental imagery
The two frequency bands play an important role.
Step 3, building Mental imagery are because imitating network；
Mental imagery is constructed because imitating network using multivariable Granger causality.The specific steps are：
3-1. determines Mental imagery because imitating network node；
The region of the corresponding electrode covering of each EEG lead is defined as a node.
3-2. uses the causality between multivariable Granger causality node metric；
The causality between each node, multivariable Granger causality are measured using multivariable Granger causality
(MVGC) it is the concept based on Granger causality, is then generalized to be mutually related X, Y set and condition multivariable situation
The interaction of lower Z.
It defines firstIndicate the vertical cascade of vector, X-Indicate the lagged variable of X, Y-Indicate the lagged variable of Y, Z-Table
Show that the lagged variable of Z, ∑ (X) indicate that the covariance matrix of n × n of X, ∑ (X, Y) indicate X, the covariance matrix of Y, tr [] table
Show the mark of matrix.
Node is predicted multiple linear regression of the nodes X on another node Y during Mental imagery task：
X=AY+ ε
Wherein A is the matrix of n × m, includes regression coefficient, and random vector ε includes residual error.The coefficient of the model passes through residual
Apply zero correlation between difference and regression forecasting factor Y to uniquely determine.By Yule-Walkers formula, we are available：
A=∑ (X, Y) ∑ (Y)-1
The covariance matrix of residual error：
∑ (ε)=∑ (X | Y)=∑ (X)-∑ (X, Y) ∑ (Y)-1∑ (X, Y)T
Assuming that we are there are three nodes X nowt, Yt, Zt.Then it is closed to measure Granger Causality of the given Z from Y to X
System, we want relatively following two multivariable autoregression MVAR model：
Therefore, it is predicted the lag that nodes X adds conditional-variable Z on self delay previous first, is then added
The lag of predictive variable Y.It is measured, i.e., is provided by regression residuals variance-rate according to the Granger causality of standard.So root
According to we formula it can be concluded that：
Predictive variable is only limitted to using the gauge of Granger causality and predictive variable Y and X are defined, because
First regression residuals variance always greater than or be equal to second people's regression residuals variance, institute's above formula always greater than be equal to 0.I
Consider now predictive variable and predictive variable not when being confined to single argument, i.e. multivariable G causality, for more
The prediction case of variable, the definition of a G causality standard not yet, a kind of possibility is simple using the equal of multivariable
Square error, the i.e. squared length of the population variance of multivariable residual error or expected polynary residual error.Therefore we it can be concluded that：
Above formula can calculate multivariable Granger causality.
3-3. is constructed according to causality because imitating network；
Each node is determined by significance test using the relationship between multivariable Granger causality quantization node
Between cause and effect stream, i.e. the size on the size of multivariable Granger causality side between node, multivariable Granger Causality
The direction on the direction of relationship side between node.
Step 4 is encoded to Mental imagery because imitating network matrix；
Because effect network matrix is exactly because the matrix of effect network indicates.We are given using convolutional neural networks because imitating network square
Battle array is encoded, and generates 8 measured values because imitating network by it by training convolutional neural networks.Although training process is more multiple
It is miscellaneous, but training only need to be primary all right, once completing, we can handle us by this convolutional neural networks for training
Because imitate network matrix, allow its for we because effect network matrix generate 8 measured values, this 8 measured values be exactly because effect network
Feature.Specific step is as follows：
4-1. mark is because imitating network matrix；
4-2. is with having marked because imitating network matrix training convolutional neural networks；
Training convolutional neural networks, it is ensured that the Euclidean distance between measured value generated under same movement imagination task is less than
Threshold value T, the Euclidean distance between measured value generated under different motion imagination task are greater than threshold value T；
4-3 allows convolutional neural networks to be respectively each Mental imagery task because effect network matrix generates 8 measured values.
Step 5 classifies to Mental imagery because imitating the feature of network using linear SVM；
This last step, exactly in finding database, compared with our measured values of the Mental imagery because imitating network matrix
That close Mental imagery is because imitating network matrix.We need to be to do is to, one linear SVM classifier of training, it
Can from new because obtain measurement result in effect network matrix, and find it is most matched that because imitating network matrix, classifier
It as a result is exactly most matched Mental imagery task.
The present invention has the beneficial effect that：
It need to only use Mental imagery because of an effect network matrix convolutional neural networks of training, and use 8 characteristic value training
Once linear support vector machine classifier, we can be carried out online recognition multiclass Mental imagery EEG signals.
Detailed description of the invention
Fig. 1 multiclass Mental imagery task recognition flow chart；
Specific embodiment：
The present invention is further explained in the light of specific embodiments.It is described below only as demonstration and explanation, not
It is intended that the present invention is limited in any way.
As shown in Figure 1, a kind of multiclass Mental imagery EEG signal identification method based on because imitating network, this method are specifically wrapped
Include following steps：
Step 1, the multiclass Mental imagery EEG signals that 64 leads are acquired using brain wave acquisition equipment；
Step 2 pre-processes collected EEG signals, including frequency reducing, noise reduction.
Step 3, building Mental imagery are because imitating network；
Step 4 is encoded to Mental imagery because imitating network matrix；
Step 5 classifies to Mental imagery because imitating the feature of network using linear SVM；
In the step 2, two frequency bands in each channel are extracted, are α (8-15HZ), β (16-30HZ) respectively.Because
The two frequency bands play an important role during Mental imagery.
In the step 3, Mental imagery is constructed because imitating network using multivariable Granger causality.Specific steps
For：
3-1. determines Mental imagery because imitating network node；
The region of the corresponding electrode covering of each EEG lead is defined as a node.
3-2. uses the causality between multivariable Granger causality node metric；
The causality between each node, multivariable Granger causality are measured using multivariable Granger causality
(MVGC) it is the concept based on Granger causality, is then generalized to be mutually related X, Y set and condition multivariable situation
The interaction of lower Z.
It defines firstIndicate the vertical cascade of vector, X-Indicate the lagged variable of X, Y-Indicate the lagged variable of Y, Z-Table
Show that the lagged variable of Z, ∑ (X) indicate that the covariance matrix of the nxn of X, ∑ (X, Y) indicate X, the covariance matrix of Y, tr [] table
Show the mark of matrix.
Multiple linear regression of the nodes X on another node Y is predicted during Mental imagery task：
X=AY+ ε
Wherein A is the matrix of n × m, includes regression coefficient, and random vector ε includes residual error.The coefficient of the model passes through residual
Apply zero correlation between difference and regression forecasting factor Y to uniquely determine.By Yule-Walkers formula, we are available：
A=∑ (X, Y) ∑ (Y)-1
The anti-variance matrix of residual error：
∑ (ε)=∑ (X | Y)=∑ (X)-∑ (X, Y) ∑ (Y)-1∑ (X, Y)T
Assuming that we are there are three nodes X nowt, Yt, Zt.Then it is closed to measure Granger Causality of the given Z from Y to X
System, we want relatively following two multivariable autoregression MVAR model：
Therefore, it is predicted the lag that nodes X adds conditional-variable Z on self delay previous first, is then added
The lag of predictive variable Y.It is measured, i.e., is provided by regression residuals variance-rate according to the Granger causality of standard.So root
According to we formula it can be concluded that：
Predictive variable is only limitted to using the gauge of Granger causality and predictive variable Y and X are defined, because
First regression residuals variance always greater than or be equal to second people's regression residuals variance, institute's above formula always greater than be equal to 0.I
Consider now predictive variable and predictive variable not when being confined to single argument, i.e. multivariable G causality, for more
The prediction case of variable, the definition of a G causality standard not yet, a kind of possibility is simple using the equal of multivariable
Square error, the i.e. squared length of the population variance of multivariable residual error or expected polynary residual error.Therefore we it can be concluded that：
Above formula can calculate multivariable Granger causality.
3-3. is constructed according to causality because imitating network；
Each node is determined by significance test using the relationship between multivariable Granger causality quantization node
Between cause and effect stream, i.e. the size on the size of multivariable Granger causality side between node, multivariable Granger Causality
The direction on the direction of relationship side between node.
In the step 4, to Mental imagery because effect network matrix is encoded；
Because effect network matrix is exactly because the matrix of effect network indicates.We are given using convolutional neural networks because imitating network square
Battle array is encoded, and generates 8 measured values because imitating network by it by training convolutional neural networks.Although training process is more multiple
It is miscellaneous, but training only need to be primary all right, once completing, we can handle us by this convolutional neural networks for training
Because imitate network matrix, allow its for we because effect network matrix generate 8 measured values, this 8 measured values be exactly because effect network
Feature.Specific step is as follows：
4-1. mark is because imitating network matrix；
4-2. is with having marked because imitating network matrix training convolutional neural networks；
Training convolutional neural networks, it is ensured that the Euclidean distance between measured value generated under same movement imagination task is less than
Threshold value T, the Euclidean distance between measured value generated under different motion imagination task are greater than threshold value T；
4-3 allows convolutional neural networks to be respectively each Mental imagery task because effect network matrix generates 8 measured values.
Step 5 classifies to Mental imagery because imitating the feature of network using linear SVM；
This last step, exactly in finding database, compared with our measured values of the Mental imagery because imitating network matrix
That close Mental imagery is because imitating network matrix.We need to do is to one classifier of training, it can be from new because of effect
Measurement result is obtained in network matrix, and find it is most matched that because imitating network matrix, the result of classifier is exactly most to match
Mental imagery task.
Claims (4)
1. a kind of multiclass Mental imagery EEG signal identification method based on because imitating network, it is characterised in that：It is walked including following 5
Suddenly：
Step 1, the multiclass Mental imagery EEG signals that 64 leads are acquired using brain wave acquisition equipment；
Step 2 pre-processes collected EEG signals, including frequency reducing, noise reduction；
Step 3 constructs Mental imagery because imitating network using multivariable Granger causality；
Step 4 is encoded to Mental imagery because imitating network matrix；
Step 5 classifies to Mental imagery because imitating the feature of network using linear SVM；
In the step 4, to Mental imagery because effect network matrix is encoded；Given using convolutional neural networks because imitating network
Matrix is encoded, and generates 8 measured values because imitating network by it by training convolutional neural networks, this 8 measured values are exactly
Because imitating the feature of network；Specific step is as follows：
4-1. mark is because imitating network matrix；
4-2. is with having marked because imitating network matrix training convolutional neural networks；
Training convolutional neural networks, it is ensured that the Euclidean distance between measured value generated under same movement imagination task is less than threshold value
T, the Euclidean distance between measured value generated under different motion imagination task are greater than threshold value T；
4-3. allows trained convolutional neural networks to be respectively each Mental imagery task because effect network matrix generates 8 measurements
Value.
2. a kind of multiclass Mental imagery EEG signal identification method based on because imitating network according to claim 1, special
Sign is：In the step 2, it is α (8-15HZ), β (16- respectively that frequency reducing process, which is to extract two frequency bands in each channel,
30HZ)。
3. a kind of multiclass Mental imagery EEG signal identification method based on because imitating network according to claim 1, special
Sign is：In the step 3, Mental imagery is constructed because imitating network using multivariable Granger causality；The specific steps are：
3-1. determines Mental imagery because imitating network node；
The region of the corresponding electrode covering of each EEG lead is defined as a node；
3-2. uses the causality between multivariable Granger causality node metric；
Assuming that three nodes Xs, Y, Z, wherein X expression are predicted node, consider that predictive variable and predictive variable be not confined to
In the case of single argument, that is, given multivariable G causality of the Z from Y to X is measured, the mean square error of multivariable, i.e. multivariable are used
The squared length of the population variance of residual error or expected polynary residual error, obtains：
Given multivariable Granger causality of the Z from Y to X is measured,Indicate the vertical cascade of vector, X-Table
Show the lagged variable of X, Y-Indicate the lagged variable of Y, Z-Indicate the lagged variable of Z,Indicate X pairsLinear regression residual error covariance matrix,Indicate X pairs's
The mark of covariance matrix tr [] representing matrix of linear regression residual error；
3-3. is constructed according to causality because imitating network；
It is determined between each node using the relationship between multivariable Granger causality quantization node by significance test
Cause and effect stream, i.e. the size on the size of multivariable Granger causality side between node, multivariable Granger causality
Direction side between node direction.
4. a kind of multiclass Mental imagery EEG signal identification method based on because imitating network according to claim 1, special
Sign is：Using linear SVM to Mental imagery because the feature of effect network is classified；Specially：Training one linear
Support vector machine classifier obtains measurement result from because of effect network matrix, and find it is most matched that because imitating network matrix,
The result of classifier is exactly most matching Mental imagery task.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810463020.0A CN108875580A (en) | 2018-05-15 | 2018-05-15 | A kind of multiclass Mental imagery EEG signal identification method based on because imitating network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810463020.0A CN108875580A (en) | 2018-05-15 | 2018-05-15 | A kind of multiclass Mental imagery EEG signal identification method based on because imitating network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108875580A true CN108875580A (en) | 2018-11-23 |
Family
ID=64333920
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810463020.0A Pending CN108875580A (en) | 2018-05-15 | 2018-05-15 | A kind of multiclass Mental imagery EEG signal identification method based on because imitating network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108875580A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109711383A (en) * | 2019-01-07 | 2019-05-03 | 重庆邮电大学 | Convolutional neural networks Mental imagery EEG signal identification method based on time-frequency domain |
| CN110969108A (en) * | 2019-11-25 | 2020-04-07 | 杭州电子科技大学 | Limb action recognition method based on autonomic motor imagery electroencephalogram |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102521505A (en) * | 2011-12-08 | 2012-06-27 | 杭州电子科技大学 | Brain electric and eye electric signal decision fusion method for identifying control intention |
| CN104305993A (en) * | 2014-11-12 | 2015-01-28 | 中国医学科学院生物医学工程研究所 | Electroencephalogram source localization method based on granger causality |
| CN104473636A (en) * | 2014-12-30 | 2015-04-01 | 天津大学 | Brain fatigue network analysis method based on partial orientation coherence |
| CN106691378A (en) * | 2016-12-16 | 2017-05-24 | 深圳市唯特视科技有限公司 | Deep learning vision classifying method based on electroencephalogram data |
-
2018
- 2018-05-15 CN CN201810463020.0A patent/CN108875580A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102521505A (en) * | 2011-12-08 | 2012-06-27 | 杭州电子科技大学 | Brain electric and eye electric signal decision fusion method for identifying control intention |
| CN104305993A (en) * | 2014-11-12 | 2015-01-28 | 中国医学科学院生物医学工程研究所 | Electroencephalogram source localization method based on granger causality |
| CN104473636A (en) * | 2014-12-30 | 2015-04-01 | 天津大学 | Brain fatigue network analysis method based on partial orientation coherence |
| CN106691378A (en) * | 2016-12-16 | 2017-05-24 | 深圳市唯特视科技有限公司 | Deep learning vision classifying method based on electroencephalogram data |
Non-Patent Citations (5)
| Title |
|---|
| ADAM B. BARRETT ET AL.: "Multivariate Granger Causality and Generalized Variance", 《ARXIV》 * |
| 佘青山 等: "基于感兴趣脑区 LASSO-Granger因果关系的脑电特征提取算法", 《电子与信息学报》 * |
| 曾庆山 等: "基于 CSP 与卷积神经网络算法的多类运动想象脑电信号分类", 《科学技术与工程》 * |
| 王卫星 等: "基于卷积神经网络的脑电信号上肢运动意图识别", 《浙江大学学报》 * |
| 黄登凤: "基于PDC的注意相关脑电因效网络分析", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109711383A (en) * | 2019-01-07 | 2019-05-03 | 重庆邮电大学 | Convolutional neural networks Mental imagery EEG signal identification method based on time-frequency domain |
| CN109711383B (en) * | 2019-01-07 | 2023-03-31 | 重庆邮电大学 | Convolutional neural network motor imagery electroencephalogram signal identification method based on time-frequency domain |
| CN110969108A (en) * | 2019-11-25 | 2020-04-07 | 杭州电子科技大学 | Limb action recognition method based on autonomic motor imagery electroencephalogram |
| CN110969108B (en) * | 2019-11-25 | 2023-04-07 | 杭州电子科技大学 | Limb action recognition method based on autonomic motor imagery electroencephalogram |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107753026B (en) | Intelligent shoe self-adaptive monitoring method for spinal leg health | |
| Karg et al. | Human movement analysis as a measure for fatigue: A hidden Markov-based approach | |
| KR20130050707A (en) | The apparutus and method for classify input pattern promptly using artificial neural network | |
| Djamal et al. | Identification of post-stroke EEG signal using wavelet and convolutional neural networks | |
| Thenmozhi et al. | Feature selection using extreme gradient boosting Bayesian optimization to upgrade the classification performance of motor imagery signals for BCI | |
| Held et al. | Extracting fuzzy rules from polysomnographic recordings for infant sleep classification | |
| CN109431492A (en) | ECG lead signals based on neural network algorithm simulate method for reconstructing | |
| CN110866537B (en) | Brain wave-based emotion recognition method for game evaluation | |
| Lei et al. | Training prediction and athlete heart rate measurement based on multi-channel PPG signal and SVM algorithm | |
| CN108875580A (en) | A kind of multiclass Mental imagery EEG signal identification method based on because imitating network | |
| CN107808166A (en) | The myoelectricity feature extracting method that a kind of MEMD tensors linear Laplace differentiates | |
| Gharaei et al. | Optimizing the setting of medical interactive rehabilitation assistant platform to improve the performance of the patients: A case study | |
| Zhang et al. | A hybrid model for blood pressure prediction from a PPG signal based on MIV and GA-BP neural network | |
| CN105769186A (en) | Complex network based electromyographic signal acquisition position selecting method | |
| KR20200063337A (en) | Apparatus and method for learning user's operation intention using artificial neural network | |
| Courtney et al. | Predicting navigation performance with psychophysiological responses to threat in a virtual environment | |
| Lv et al. | Prediction of hand grip strength based on surface electromyographic signals | |
| Kim et al. | Development of person-independent emotion recognition system based on multiple physiological signals | |
| CN116831874A (en) | Lower limb rehabilitation device control method based on electromyographic signals | |
| Chen et al. | Research on EEG classification with neural networks based on the Levenberg-Marquardt algorithm | |
| Zhang et al. | [Retracted] Influence of Sports Biomechanics on Martial Arts Sports and Comprehensive Neuromuscular Control under the Background of Artificial Intelligence | |
| CN115736840A (en) | Sleep quality identification and classification method based on electrocardiogram data | |
| CN114587391A (en) | Brain-computer interface-based rehabilitation training device and training method | |
| Jaimes-Albarracin et al. | Brain and heart physiological networks analysis employing neural networks granger causality | |
| Sullivan et al. | Reconstruction of missing physiological signals using artificial neural networks |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181123 |
|
| RJ01 | Rejection of invention patent application after publication |