Please wait a minute...
Shopping Cart Email List Chinese
Advanced Search Fig/Tab
Frontiers of Computer Science

ISSN 2095-2228

ISSN 2095-2236(Online)

CN 10-1014/TP

Postal Subscription Code 80-970

2018 Impact Factor: 1.129

Featured Articles  |  Most Downloaded  |  Most Reading
Online Submission Subscribe to Our RSS Content Feed
Front. Comput. Sci.    2022, Vol. 16 Issue (5) : 165328    https://doi.org/10.1007/s11704-021-0587-2
RESEARCH ARTICLE
Multiband decomposition and spectral discriminative analysis for motor imagery BCI via deep neural network
Pengpai WANG, Mingliang WANG, Yueying ZHOU, Ziming XU, Daoqiang ZHANG()
College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, MIIT Key Laboratory of Pattern Analysis and Machine Intelligence, Nanjing 211106, China
 Download: PDF(13632 KB)   HTML
 Export: BibTeX | EndNote | Reference Manager | ProCite | RefWorks
Abstract

Human limb movement imagery, which can be used in limb neural disorders rehabilitation and brain-controlled external devices, has become a significant control paradigm in the domain of brain-computer interface (BCI). Although numerous pioneering studies have been devoted to motor imagery classification based on electroencephalography (EEG) signal, their performance is somewhat limited due to insufficient analysis of key effective frequency bands of EEG signals. In this paper, we propose a model of multiband decomposition and spectral discriminative analysis for motor imagery classification, which is called variational sample-long short term memory (VS-LSTM) network. Specifically, we first use a channel fusion operator to reduce the signal channels of the raw EEG signal. Then, we use the variational mode decomposition (VMD) model to decompose the EEG signal into six band-limited intrinsic mode functions (BIMFs) for further signal noise reduction. In order to select discriminative frequency bands, we calculate the sample entropy (SampEn) value of each frequency band and select the maximum value. Finally, to predict the classification of motor imagery, a LSTM model is used to predict the class of frequency band with the largest SampEn value. An open-access public data is used to evaluated the effectiveness of the proposed model. In the data, 15 subjects performed motor imagery tasks with elbow flexion / extension, forearm supination / pronation and hand open/close of right upper limb. The experiment results show that the average classification result of seven kinds of motor imagery was 76.2%, the average accuracy of motor imagery binary classification is 96.6% (imagery vs. rest), respectively, which outperforms the state-of-the-art deep learning-based models. This framework significantly improves the accuracy of motor imagery by selecting effective frequency bands. This research is very meaningful for BCIs, and it is inspiring for end-to-end learning research.
Keywords brain computer interface      EEG      long short-term memory      VMD      sample entropy      motor imagery     
Corresponding Author(s): Daoqiang ZHANG   
Just Accepted Date: 30 April 2021   Issue Date: 24 December 2021
 Cite this article:   
Pengpai WANG,Mingliang WANG,Yueying ZHOU, et al. Multiband decomposition and spectral discriminative analysis for motor imagery BCI via deep neural network[J]. Front. Comput. Sci., 2022, 16(5): 165328.
 URL:  
https://academic.hep.com.cn/fcs/EN/10.1007/s11704-021-0587-2
https://academic.hep.com.cn/fcs/EN/Y2022/V16/I5/165328
Fig.1  VS-LSTM model structure. Participants perform six kinds of motor imagery (MI) and obtain classification results through five steps. (a) EEG data collection and pretreatment. Participants wear EEG equipment for MI and collect data to obtain 64-channel EEG data. A series of data pre-processing operations are executed, including (1) select channels, (2) band pass filter, (3) re-reference, (4) segment epoch, (5) ICA analysis, etc.; (b) The channel fusion of EEG. We use channel fusion operator to calculate EEG channels through weighted average operator, maximum operator and minimum operator, which fuse 3D data into 2D; (c) EEG mode decomposition of VMD. Variational mode decomposition (VMD) model is employed to decompose EEG into six band-limited intrinsic mode functions (BIMFs); (d) Calculation of sample entropy. Sample entropy is utilized to calculate each BIMF and the max value is selection; (e) Classification of LSTM model. We employ LSTM to classify the BIMF frequency band, which is the max value of sample entropy. Finally, we obtain the classification result of each MI
Fig.2  EEG Signal decomposed by VMD to six BIMFs with different frequency band and threshold
Fig.3  Channel locations of 61 EEG electrode. The upper part is the frontal lobe and the lower part is the occipital lobe
Fig.4  Experimental paradigm of motor imagery. Every participant was asked to perform 10 rounds, 6 MI classes and a resting state category and recorded 42 trials per run. In single trial, a cross appeared together with a beep sound in screen at second 0. After 2 seconds, the subjects imagined a continuous movement while appearing a prompt. In the end of the exercise, a break with a random duration of 2s to 3s was followed
Fig.5  Location of datasets 2a (22 channels) and 2b (three channels)
Fig.6  Binary classification of six MI classes with rest class in every BIMF, every subgraph is the classifications between six MIs and rest class of 15 subjects. (a) BIMF 1; (b) BIMF 2; (c) BIMF 3; (d) BIMF 4; (e) BIMF 5; (f) BIMF 6
Subjects IV_2a/% IV_2b/%
1 2 3 4 5 6
S1 92.6 94.5 94.6 95.6 94.5 96.8 97.3
S2 93.5 95.3 94.9 97.2 95.8 97 96.9
S3 91.3 93.6 95.7 94.7 94.9 96.8 95.1
S4 96.7 97.1 96.8 98.3 98.1 95.2 98.4
S5 89.8 91.7 89.6 93.5 92.8 94.9 96.3
S6 86.4 94.9 95.2 88.7 96.4 98.2 97.4
S7 92.4 92.5 94.1 94.5 95.8 96.3 96.8
S8 95.8 96.3 93.7 92.7 94.7 98.2 95.5
S9 94.2 96.6 94.5 93.6 95 94.6 97.2
Average 92.52 94.72 94.34 94.31 95.33 96.44 96.77
Tab.1  Binary classification of MI in two datasets
Subjects DSP-LDA/% LSTM/% CNN/% VS-LSTM (Ours)/%
S1 71.4 93.2 90 96.4
S2 72.3 86.9 86.3 96.8
S3 68.7 91.7 88.5 96.8
S4 78.6 88.7 86.9 96.7
S5 77.3 91.9 91.4 96.8
S6 81.3 92.9 92.2 96.4
S7 66.6 97 86.1 96.4
S8 85.1 95 94.3 95.7
S9 68.9 96.2 96.4 96.9
S10 66.7 89.7 92.4 96.1
S11 77.5 92.2 94.1 96.7
S12 66.9 86.8 91.6 97.2
S13 74.4 91.6 88.2 96.8
S14 73.5 95.9 94.8 96.4
S15 76.8 89.8 92.3 97.2
Average 73.7 92.1 91 96.6
Tab.2  Average classification accuracy of four models
Subjects 1 vs. 2/% 1 vs. 3/% 1 vs. 4/% 1 vs. 5/% 1 vs. 6/% 2 vs. 3/% 2 vs. 4/% 2 vs. 5/% 2 vs. 6/% 3 vs. 4/% 3 vs. 5/% 5 vs. 6/%
S1 91.4 93.2 90 96.4 92.5 90.6 93.4 90 86.4 92.5 90.6 93.4
S2 93.5 86.9 86.3 96.8 93.4 86.7 95.2 86.3 96.8 93.4 86.7 95.2
S3 88.7 91.7 88.5 96.8 90.5 82.5 95.4 99.5 96.6 90.5 82.5 95.4
S4 89.6 88.7 89 96.7 94.6 92.1 93.2 86.9 96.7 94.6 92.1 93.2
S5 93.3 91.9 91.4 96.8 91.2 97.3 96.4 91.4 96.8 91.2 94.3 96.4
S6 92.3 92.9 92.2 96.4 90.8 91.6 91.2 92.2 96.4 90.8 91.6 97.2
S7 87.6 97 86.1 96.4 92.6 89.2 86.7 86.1 96.4 92.6 89.2 86.7
S8 94.1 95 94.3 95.7 94.7 84.7 91 94.3 95.7 94.7 94.7 89.2
S9 96.9 96.2 96.4 96.9 95.2 96.3 86.7 96.4 96.9 95.2 96.3 86.7
S10 86.7 89.7 92.4 96.1 93 97.2 86.5 92.4 96.1 93 87.2 96.5
S11 97.5 92.2 94.1 96.7 91.4 86.3 89.2 94.1 96.7 91.4 86.3 89.2
S12 96.9 86.8 91.6 97.2 93.6 94.2 93.5 91.6 97.2 93.6 94.2 93.5
S13 94.3 91.6 88.2 96.8 90.4 93.6 97.3 98.2 96.8 90.4 93.6 94.3
S14 93.7 95.9 94.8 96.4 89.5 94.2 96.2 94.8 96.4 89.5 96.2 96.2
S15 96.8 89.8 92.3 97.2 88.4 94.1 93.6 92.3 97.2 88.4 94.1 93.6
Average 92.89 91.97 91.17 96.62 92.12 91.37 92.37 92.43 95.94 92.12 91.31 93.11
Tab.3  Binary classification accuracy of six MI
Model Accuracy /%
DSP-LDA 25
DSP-SVM 36.5
LSTM 67.9
CNN 65.9
VS-LSTM (Ours) 76.2
Tab.4  Six multi-category classification of MI
Fig.7  Confusion matrix of MI and rest classes
BIMFs MI 1 MI 2 MI 3 MI 4 MI 5 MI 6 Average
BIMF 1 20.4 19.55 15.82 14.3 14.41 16.72 16.87
BIMF 2 38.7 38.79 37.23 39.3 37.45 38.44 38.32
BIMF 3 38.8 38.67 38.63 41 37.82 38.76 38.95
BIMF 4 33.4 32.79 34.4 37.4 32.84 33.83 34.11
BIMF 5 33.1 33.27 34.01 35 36.4 35.21 34.5
BIMF 6 37.9 37.92 37.61 38.6 37.17 37.64 37.81
Average 33.72 33.5 32.95 34.27 32.68 33.43 33.43
Tab.5  Six sample entropy values in BIMFs
K MI 1 MI 2 MI 3 MI 4 MI 5 MI 6 Average
1 62.3 58.68 64.21 64.7 54.51 66.24 66.30
2 68.57 68.51 67.29 66.31 61.46 64.62 68.54
3 67.8 68.67 68.63 77.14 77.22 68.78 68.76
4 83.4 82.79 84.4 87.4 82.84 83.83 74.18
5 93.1 83.27 94.01 85.17 86.82 95.28 84.55
6 97.87 92.52 94.61 98.64 97.65 97.63 96.57
7 83.21 83.69 92.57 84.27 92.67 93.63 93.43
8 87.54 84.64 78.69 84.92 91.54 93.28 91.54
9 84.65 81.12 76.54 81.57 86.25 90.22 86.51
10 80.97 76.87 72.05 70.84 74.88 76.98 72.54
11 78.40 75.41 70.6 65.84 73.84 72.45 70.52
12 75.4 70.18 60.54 64.25 68.74 67.23 64.0
Tab.6  Binary classification of different K values
MI class VMD-SE-CNN/% VMD-LSTM/% SE-CNN/% SE-LSTM/% LSTM/% VS-LSTM(Ours)/%
MI 1 93.4 94.8 88.7 91.3 87.4 98.7
MI 2 92.8 93.1 84.6 90.4 86.3 96.9
MI 3 90.9 92.7 82.1 85.7 87.5 94.5
MI 4 89.3 90.7 83.7 86.1 86.1 95.8
MI 5 92.5 91.5 84.9 89.7 87.5 97.7
MI 6 90.4 92.4 87.6 87.6 86.2 96.2
Average 91.6 92.5 85.3 88.5 86.8 96.6
Tab.7  MI class average accuracy of classification with10-fold cross-validation
Fig.8  Subject average accuracy of classification with10-fold cross-validation
Model Accuracy/%
VMD-SE-CNN 62.6
VMD-LSTM 58.7
SE-CNN 64.2
SE-LSTM 65.9
VS-LSTM (Ours) 76.2
Tab.8  Six multi-category classification of MI
Fig.9  An example of t-SNE visualization for raw data and out of layer from S3. Red, purple, green, orange, yellow and blue points represent elbow flexion / extension, forearm reinforcement / pronation and hand open / close states respectively. (a) Raw date; (b) output of layer
Fig.10  Convergence curves of binary classification of VS-LSTM algorithms
Fig.11  Computational time by each of the compared algorithms for Ml EEG classification
1 Y Li , J Pan , F Wang , Z Yu . A hybrid BCI system combining P300 and SSVEP and its application to wheelchair control. IEEE Transactions on Biomedical Engineering, 2013, 60( 11): 3156– 3166
2 L F Nicolas , J Gomez . Brain computer interfaces, a review. Sensors, 2012, 12 : 1211– 1279
3 J R Wolpaw , R S Bedlack , D J Reda , R J Ringer , P G Banks , T M Vaughan , S M Heckman , L M McCane , C S Carmack , S Winden , D J McFarland , E W Sellers , H R Shi , T Paine , D S Higgins , A C Lo , H S Patwa , K J Hill , G D Huang , R L Ruff . Independent home use of a brain-computer interface by people with amyotrophic lateral sclerosis. Neurology, 2018, 91( 3): 258– 267
4 D Delisle , V Cardoso , D Gurve , F Loterio , M A Romero-Laiseca , S Krishnan , T Bastos . System based on subject-specific bands to recognize pedaling motor imagery: towards a BCI for lower-limb rehabilitation. Journal of Neural Engineering, 2019, 16( 5): 056005–
5 J Wilmskoetter , J D Gaizo , L Phillip , R Behroozmand , E Gleichgerrcht , J Fridriksson , E Riley , L Bonilha . Predicting naming responses based on pre-articulatory electrical activity in individuals with aphasia. Clinical Neurophysiology, 2019, 130( 11): 2153– 2163
6 C Yang , H Zhang , S Zhang , X Han , S Gao , X Gao . The spatio-temporal equalization for evoked or event-related potential detection in multichannel EEG data. IEEE Transaction on Biomedical Engineering, 2020, 67 : 2397– 2414
7 M Xu , X Xiao , Y Wang , H Qi , T P Jung , D Ming . A brain-computer interface based on miniature-event-related potentials induced by very small lateral visual stimuli. IEEE Transaction on Biomedical Engineering, 2018, 65 : 1166– 1175
8 R Ramele , A J Villar , J M Santos . EEG waveform analysis of p300 ERP with applications to brain computer interfaces. Brain Sciences, 2018, 8( 11): 199–
9 M Nakanishi , Y Wang , X Chen , Y T Wang , X Gao , T P Jung . Enhancing detection of SSVEPs for a high-speed brain speller using task-related component analysis. IEEE Transaction on Biomedical Engineering, 2018, 65 : 104– 112
10 M Nakanishi , Y Wang , X Chen , Y T Wang , X Gao , T P Jung . Enhancing detection of SSVEPs for a high-speed brain speller using task-related component analysis. IEEE Transaction on Biomedical Engineering, 2018, 65( 1): 104– 112
11 X Han , K Lin , S Gao , X Gao . A novel system of SSVEP-based human-robot coordination. Journal of Neural Engineering, 2019, 16( 1): 016006–
12 A M Azab , L Mihaylova , K K Ang , M Arvaneh . Weighted transfer learning for improving motor imagery-based brain-computer interface. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2019, 27 : 1352– 1359
13 Y Jiao , Y Zhang , X Chen , E Yin , J Jin , X Wang , A Cichocki . Sparse group representation model for motor imagery EEG classification. IEEE Journal of Biomedical and Health Informatics, 2019, 23 : 631– 641
14 Y Zhang , Y Wang , J Jin , X Wang . Sparse Bayesian learning for obtaining sparsity of EEG frequency bands based feature vectors in motor imagery classification. International Journal of Neural Systems, 2017, 27( 2): 1650032–
15 Y R Tabar , U Halici . A novel deep learning approach for classification of EEG motor imagery signal. Journal of Neural Engineering, 2017, 14( 1): 016003–
16 B J Edelman , B Baxter , B He . EEG source imaging enhances the decoding of complex right-hand motor imagery tasks. IEEE Transaction on Biomedical Engineering, 2016, 63 : 4– 14
17 E Quiles , S Ferran , G Candela . Low-cost robotic guide based on a motor imagery brain-computer interface for arm assisted rehabilitation. International Journal of Environmental Research and Public Health, 2020, 17( 3): 699–
18 R Mane , E Chew , K S Phua , K A Kai , C Guan . Prognostic and monitory EEG-biomarkers for BCI upper-limb stroke rehabilitation. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2019, 27 : 1654– 1664
19 A Nourmohammadi , M Jafari , T Zander . A survey on unmanned aerial vehicle remote control using brain-computer Interface. IEEE Transaction on Human-Machine Systems, 2018, 48( 4): 337– 348
20 A Jafarifarmand , M. A Badamchizadeh . EEG artifacts handling in a real practical brain-computer interface controlled vehicle. IEEE Transaciton on Neural Systems and Rehabilitation Engineering, 2019, 27( 6): 1200– 1208
21 B J Edelman , J Meng , D Suma , C Suma , E Zurn . Noninvasive neuroimaging enhances continuous neural tracking for robotic device control. Science Robotics, 2019, 4(31): eaaw6844 :
22 Y Su , W Li , N Bi , Z Lv . Adolescents environmental emotion perception by integrating EEG and eye movements. Frontiers in Neurorobotics, 2019, 13 : 46–
23 W Klimesch . Alpha-band oscillations, attention, and controlled access to stored information. Trends in Cognitive Sciences, 2012, 16 : 606– 617
24 G Bartur , H Pratt , N Soroker . Changes in mu and beta amplitude of the EEG during upper limb movement correlate with motor impairment and structural damage in subacute stroke. Clinical Neurophysiology, 2019, 130 : 1644– 1651
25 M Uji , R Wilson , S T Francis , K J Mullinger , S D Mayhew . Exploring the advantages of multiband fMRI with simultaneous EEG to investigate coupling between gamma frequency neural activity and the BOLD response in humans. Human Brain Mapping, 2018, 39 : 1673– 1687
26 P Milz , R D Pascual-Marqui , P Achermann , K Kochi , P L Faber . The EEG microstate topography is predominantly determined by intracortical sources in the alpha band. Neuroimage, 2017, 162 : 353– 361
27 Y Zhao , Y Zhao , D Pholpat , L Chen , P G Sarrigiannis . Imaging of nonlinear and dynamic functional brain connectivity based on EEG recordings with the application on the diagnosis of alzheimer's disease. IEEE Transaction on Medical Imaging, 2020, 39 : 1571– 1581
28 Q Li , D Şentürk , C A Sugar , S Jeste , C DiStefano , J Frohlich , D Telesca . Inferring brain signals synchronicity from a sample of EEG readings. Journal of the American Statistical Association, 2019, 114 : 991– 1001
29 J Jiang , Z Yan , T Shen , G Xu , Q Guan , Z Yu . Use of deep belief network model to discriminate mild cognitive impairment and normal controls based on EEG, eye movement signals and neuropsychological tests. Journal of Medical Imaging and Health Informatics, 2019, 9 : 1978– 1985
30 X Chen , Q Chen , Y Zhang , Z J Wang . A novel EEMD-CCA approach to removing muscle artifacts for pervasive EEG. IEEE Sensors Journal, 2019, 19 : 8420– 8431
31 D Zhang , L Yao , K Chen , S Wang , X Chang , Y Liu . Making sense of Spatio-temporal preserving representations for EEG-based human intention recognition. IEEE Transaction on Cybernetics, 2019, 50 : 3033– 3044
32 D Aline , R Delorme , C Delanoë , F Amsellem , A Beggiato , D Germanaud , T Bourgeron , R Toro , G Dumas . Alpha waves as a neuromarker of autism spectrum disorder: the challenge of reproducibility and heterogeneity. Frontiers in Neuroscience, 2018, 12 : 662–
33 Y Su , P Chen , X Liu , W Li , Z Lv . A spatial filtering approach to environmental emotion perception based on electroencephalography. Medical Engineering & Physics, 2018, 60 : 77– 85
34 A Ghaemi , E Rashedi , A M Pourrahimi , M Kamandar , F Rahdari . Automatic channel selection in EEG signals for classification of left or right hand movement in Brain Computer Interfaces using improved binary gravitation search algorithm. Biomedical Signal Processing and Control, 2017, 33 : 109– 118
35 H K Lee , Y S Choi . Application of continuous wavelet transform and convolutional neural network in decoding motor imagery brain-computer interface. Entropy, 2019, 21( 12): 1199–
36 W Zeng , M Li , C Yuan , Q Wang , F Liu , Y Wang . Identification of epileptic seizures in EEG signals using time-scale decomposition (ITD), discrete wavelet transform (DWT), phase space reconstruction (PSR) and neural networks. Artificial Intelligence Review, 2020, 53 : 3059– 3088
37 S H Oh . Improving the subject independent classification of implicit intention by generating additional training data with PCA and ICA. International Journal of Contents, 2018, 14 : 24– 29
38 J Jin , R Xiao , I Daly , Y Miao , A Cichocki . Internal feature selection method of CSP based on L1-norm and dempster-shafer theory. IEEE Transactions on Neural Networks and Learning Systems, 2020,
https://doi.org/10.1109/TNNLS.2020.3015505
39 B T Sachin , V Sharma , D Siuly , S Sengur . Features based on analytic IMF for classifying motor imagery EEG signals in BCI applications. Measurement, 2018, 116 : 68– 76
40 P Balaiah . Comparative evaluation of adaptive filter and neuro-fuzzy filter in artifacts removal from electroencephalogram signal. American Journal of Applied Sciences, 2012, 9( 10): 1583– 1593
41 R Ameri , A Pouyan , V Abolghasemi . Projective dictionary pair learning for EEG signal classification in brain computer interface applications. Neurocomputing, 2016, 218 : 382– 389
42 J Ma , D Zhang . EEG signals feature extraction based on DWT and EMD combined with approximate entropy. Brain Sciences, 2019, 9( 8): 201–
43 J Jin , Y Miao , I Daly , C Zuo , A Cichocki . Correlation-based channel selection and regularized feature optimization for MI-based BCI. Neural Networks, 2019, 118 : 262– 270
44 X Zhu , P Li , C Li . Separated channel convolutional neural network to realize the training free motor imagery BCI systems. Biomedical Signal Processing and Control, 2019, 49 : 396– 403
45 B Alchalabi , J Faubert . A comparison between BCI simulation and neurofeedback for forward/backward navigation in virtual reality. Computational Intelligence and Neuroscience, 2019, 4 : 1– 12
46 M Aljal , D Ridha , I Sutrisno . Robot navigation using a brain computer interface based on motor imagery. Journal of Medical and Biological Engineering, 2019, 39( 4): 508– 522
47 L Huang , Y Zhao , Y Zeng , Z Lin . BHCR: RSVP target retrieval BCI framework coupling with CNN by a Bayesian method. Neurocomputing, 2017, 238 : 255– 268
48 Y Zhang , T Zhou , W Wu , H Xie , A Cichocki . Improving EEG decoding via clustering-based multi-task feature learning. IEEE Transactions on Neural Networks and Learning Systems, 2021,
https://doi.org/10.1109/TNNLS.2021.3053576
49 Y Jiao , T Zhou , L Yao , G Zhou , Y Zhang . Multi-view multi-scale optimization of feature representation for EEG classification improvement. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2020, 28( 12): 2589– 2597
50 X Zhang, L Yao, X Wang, J Monaghan, D Mcalpine, Y Zhang. A survey on deep learning-based non-invasive brain signals: recent advances and new frontiers. Journal of Neural Engineering, 2021, 18(3): 1−44
51 A Craik , Y He , L Jose . Contreras-vidal deep learning for electroencephalogram (EEG) classification tasks: a review. Journal of Neural Engineering, 2019, 16( 3): 031001–
52 M Wang , C Lian , D. Shen D Yao . Spatial-temporal dependency modeling and network hub detection for functional MRI analysis via convolutional-recurrent network. IEEE Transactions on Biomedical Engineering, 2020, 67( 8): 2241– 2252
53 U Acharya , S L Oh , Y Hagiwara , J H Tan , H Adeli . Deep convolutional neural network for the automated detection and diagnosis of seizure using EEG signals. Computers in Biology and Medicine, 2018, 100 : 270– 278
54 Y Hou , L Zhou , S Jia , X Lun . A novel approach of decoding EEG four-class motor imagery tasks via scout ESI and CNN. Journal of Neural Engineering, 2019, 17( 1): 016048–
55 J Chen , Z Yu , Z Gu , Y Li . Deep temporal-spatial feature learning for motor imagery-based brain-computer interfaces. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2020, 28( 11): 2356– 2366
56 Y Li , X R Zhang , B Zhang , M Y Lei , W G Cui , Y Z Guo . A channel-projection mixed-scale convolutional neural network for motor imagery EEG decoding. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2019, 27( 6): 1170– 1180
57 X Ma , S Qiu , W Wei , S Wang , H He . Deep channel-correlation network for motor imagery decoding from the same limb. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2020, 28( 1): 297– 306
58 T J Luo , C L Zhou , F Luo . Exploring spatial-frequency-sequential relationships for motor imagery classification with recurrent neural network. BMC Bioinformatics, 2018, 19(1): 1−18 :
59 P Wang , A Jiang , X Liu , J Shang , L Zhang . LSTM-based EEG classification in motor imagery tasks. IEEE Transactions Neural Systems and Rehabilitation Engineering, 2018, 26( 11): 2086– 2095
60 Z Tayeb , J Fedjaev , N Ghaboosi , C Richter , L Everding , X Qu , Y Wu , G Cheng , J Conradt . Validating deep neural networks for online decoding of motor imagery movements from EEG signals. Sensors, 2019, 19( 1): 210–
61 P Ofner , A Schwarz , J Pereira , G R Müller-Putz . Upper limb movements can be decoded from the time-domain of low-frequency EEG. PLoS ONE, 2017, 12( 8): e0182578–
62 M Tangermann , K R Müller , A Aertsen , N Birbaumer , B Blankertz . Review of the BCI competition IV. Frontiers on Neuroscience, 2012, 6 : 55–
63 N U Rehman , H Aftab . Multivariate variational mode decomposition. IEEE Transaction on Signal Processing, 2019, 67 : 6039– 6052
64 Q Cheng , W Yang , K Liu , W Zhao , Y Yang . Increased sample entropy in EEGs during the functional rehabilitation of an injured brain. Entropy, 2019, 21( 7): 698–
65 C V Martinez , V E Santamaria , R Hornero . Asynchronous control of P300-based brain-computer interfaces using sample entropy. Entropy, 2019, 21( 3): 230–
66 M Wang , D Zhang , J Huang , P T Yap , M Liu . Identifying autism spectrum disorder with multi-site fMRI via low-rank domain adaptation. IEEE Transactions on Medical Imaging, 2020, 39( 3): 644– 655
67 J H Fan , C L Sun , C Chen , J Chen , X Xinyu Liu , X Zhao , L Meng , C Dai , W Chen . EEG data augmentation: towards class imbalance problem in sleep staging tasks. Journal of Neural Engineering, 2020, 17( 5): 056017–
[1] Yujin CHAI, Yanlin WENG, Lvdi WANG, Kun ZHOU. Speech-driven facial animation with spectral gathering and temporal attention[J]. Front. Comput. Sci., 2022, 16(3): 163703-.
[2] Dongjin YU, Lin WANG, Xin CHEN, Jie CHEN. Using BiLSTM with attention mechanism to automatically detect self-admitted technical debt[J]. Front. Comput. Sci., 2021, 15(4): 154208-.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed