EEGNet: a compact convolutional neural network for EEG-based brain-computer interfaces
Open accessThe DOI matches its Crossref record; 113 of the 113 references it cites were found in a registry or an index; 18 of the 30 sources its claims cite were checked for support.
Abstract
OBJECTIVE: Brain-computer interfaces (BCI) enable direct communication with a computer, using neural activity as the control signal. This neural signal is generally chosen from a variety of well-studied electroencephalogram (EEG) signals. For a given BCI paradigm, feature extractors and classifiers are tailored to the distinct characteristics of its expected EEG control signal, limiting its application to that specific signal. Convolutional neural networks (CNNs), which have been used in computer vision and speech recognition to perform automatic feature extraction and classification, have successfully been applied to EEG-based BCIs; however, they have mainly been applied to single BCI paradigms and thus it remains unclear how these architectures generalize to other paradigms. Here, we ask if we can design a single CNN architecture to accurately classify EEG signals from different BCI paradigms, while simultaneously being as compact as possible. APPROACH: In this work we introduce EEGNet, a compact convolutional neural network for EEG-based BCIs. We introduce the use of depthwise and separable convolutions to construct an EEG-specific model which encapsulates well-known EEG feature extraction concepts for BCI. We compare EEGNet, both for within-subject and cross-subject classification, to current state-of-the-art approaches across four BCI paradigms: P300 visual-evoked potentials, error-related negativity responses (ERN), movement-related cortical potentials (MRCP), and sensory motor rhythms (SMR). MAIN RESULTS: We show that EEGNet generalizes across paradigms better than, and achieves comparably high performance to, the reference algorithms when only limited training data is available across all tested paradigms. In addition, we demonstrate three different approaches to visualize the contents of a trained EEGNet model to enable interpretation of the learned features. SIGNIFICANCE: Our results suggest that EEGNet is robust enough to learn a wide variety of interpretable features over a range of BCI tasks. Our models can be found at: https://github.com/vlawhern/arl-eegmodels.
Overview
CoChat Summary
TL;DR
EEGNet is a compact CNN using depthwise and separable convolutions that generalizes across P300, ERN, MRCP, and SMR BCI paradigms while using far fewer parameters than reference CNNs. It achieves comparable or better performance than state-of-the-art approaches in many within- and cross-subject comparisons and produces neurophysiologically interpretable features.
Key findings
- EEGNet models were up to two orders of magnitude smaller than DeepConvNet and ShallowConvNet: e.g., P300 EEGNet-4,2 had 1,066 parameters vs. 174,127 for DeepConvNet and 104,002 for ShallowConvNet; SMR EEGNet-4,2 had 796 vs. 152,219 and 40,644, respectively.
- Within-subject, EEGNet-4,2 and EEGNet-8,2 did not differ significantly across paradigms (p > 0.05). Both EEGNet models significantly outperformed all other models on MRCP (p < 0.05); EEGNet-8,2 outperformed DeepConvNet, ShallowConvNet, and xDAWN+RG on ERN (p < 0.05); all CNN models outperformed xDAWN+RG on P300 (p < 0.05).
- Cross-subject, all CNN models outperformed xDAWN+RG on P300 (p < 0.05); EEGNet-8,2 and DeepConvNet outperformed ShallowConvNet on MRCP (p < 0.05); xDAWN+RG outperformed all CNN models on ERN (p < 0.05); no significant CNN differences were found for SMR (p > 0.05).
- Feature analyses showed interpretable structure: P300 ablation gave full-model AUC 0.9054, with removal of filters (1,3,4) reducing AUC to 0.7108; SMR spatial filters learned by EEGNet-8,2 correlated with FBCSP filters (e.g., ρ = 0.93, ρ = 0.83, and ρ = −0.93).
Methods
- EEGNet uses temporal convolution, depthwise convolution for frequency-specific spatial filters, and separable convolution, with batch normalization, ELU, dropout, average pooling, and softmax classification; trained with Adam, 500 epochs, validation stopping.
- Evaluated on four datasets: P300 (15 subjects, ~2000 trials/subject, 1–40 Hz), ERN (26 subjects, 340 trials/subject, 1–40 Hz), MRCP (13 subjects, ~1100 trials/subject, 0.1–40 Hz), and SMR (9 subjects, 288 trials/subject, 4–40 Hz).
- Within-subject used four-fold blockwise cross-validation; cross-subject used repeated random folds (30 for P300/MRCP/ERN, 90 for SMR). Comparisons included DeepConvNet, ShallowConvNet, xDAWN+RG for ERP datasets, and FBCSP for SMR, with repeated-measures or one-way ANOVA.
Limitations
- The paper does not state a formal limitations section. Evident from the text: validation is limited to four BCI paradigms/datasets, and cross-subject ERN results show xDAWN+RG significantly outperforming all CNN models.
- The comparison to xDAWN+RG used a single model rather than the original ensemble and omitted dataset-specific “meta features,” as stated by the authors.
- The authors note that ShallowConvNet’s severe underperformance on within-subject MRCP was unexpected and that determining its underlying source will be explored in future research.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
8 Valid · 1 Not valid · 2 Not checked · 61 Own results and interpretations
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“each paradigm relies on manual specification of signal processing [7], feature extraction [8] and classification methods [9], a process which often requires significant subject-matter expertise and/or a priori knowledge about the expected EEG signal.”
What the sources say
- Partly supportedcites [7] [7] A. Bashashati, M. Fatourechi, R. K. Ward, and G. E. Birch, 'A survey of signal processing algorithms in brain-computer interfaces based on electrical brain signals,' Journal of Neural Engineering , vol. 4, no. 2, p. R32, 2007. [Online]. Available: http://stacks.iop.org/1741-2552/4/i=2/a=R03 - sourceExperimental — accuracy not yet measured
For the part this citation carries (manual specification of the signal processing), our copy supports the substance: the survey's Section 5.1 (signal enhancement) states that 'the choice of a suitable pre-processing technique however is dependent on several factors such as the recording technology, number of electrodes, and neuromechanism of the BCI', and Section 3 states that 'feature extraction methods used in BCI systems are closely related to the specific neuromechanism(s) used by a BCI' (e.g. VEP-based BCIs extract features to detect the evoked potential in the ongoing EEG), and Section 5.2.2 notes parametric modeling 'require[s] an a priori choice of the structure and order of the signal generation mechanism model'. However, no passage in our copy states that this selection process requires 'significant subject-matter expertise' — the paper frames itself as a reference/taxonomy of methods used (Introduction; Conclusions) rather than characterizing the expertise demand. The expertise clause of the claim is therefore not supported by this source. (LLM, conf=0.72)
- Not checked — no full text of the sourcecites [8] [8] D. J. McFarland, C. W. Anderson, K. R. Muller, A. Schlogl, and D. J. Krusienski, 'Bci meeting 2005-workshop on bci signal processing: feature extraction and translation,' IEEE Transactions on Neural Systems and Rehabilitation Engineering , vol. 14, no. 2, pp. 135-138, June 2006. - source
- Partly supportedcites [9] [9] F. Lotte, M. Congedo, A. L´ ecuyer, F. Lamarche, and B. Arnaldi, 'A review of classification algorithms for eeg-based brain-computer interfaces,' Journal of Neural Engineering , vol. 4, no. 2, p. R1, 2007. [Online]. Available: http://stacks.iop.org/1741-2552/4/i=2/a=R01 - sourceExperimental — accuracy not yet measured
Section 2.2 says that choosing an appropriate classifier requires knowing the properties of available classifiers, and Section 4 provides guidelines for choosing one for a given BCI context. This supports the classification-method selection part attributed to [9], but the paper does not establish that every paradigm requires manual specification or that the process requires significant subject-matter expertise or prior knowledge of the expected EEG signal. Page numbers are unavailable in our copy. (LLM, conf=0.81)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
The cited sources that could be checked couldn’t confirm or dispute this claim.
In the paper
“For all ERP-based data analyses (P300, ERN, MRCP) the traditional approach is the approach which won the Kaggle BCI Competition (code and documentation at http://github.com/alexandrebarachant/bci-challenge-ner-2015), which uses a combination of xDAWN Spatial Filtering [83], Riemannian Geometry [84,85], channel subset selection and L 1 feature regularization”
What the sources say
- Could not verifycites [83] [82] J. T. Springenberg, A. Dosovitskiy, T. Brox, and M. A. Riedmiller, 'Striving for simplicity: The all convolutional net,' arXiV , vol. abs/1412.6806, 2014. [Online]. Available: http://arxiv.org/abs/1412.6806 - sourceExperimental — accuracy not yet measured
The Experiments section (p. 4) describes image-classification experiments on CIFAR-10, CIFAR-100 and ImageNet. Searches for the Kaggle BCI Competition, ERP/P300 and xDAWN returned no passage in our copy addressing the claimed approach or its components. (LLM, conf=0.78)
- Not checked — no full text of the sourcecites [84] [83] B. Rivet, A. Souloumiac, V. Attina, and G. Gibert, 'xDAWN algorithm to enhance evoked potentials: Application to brain-computer interface,' IEEE Transactions on Biomedical Engineering , vol. 56, no. 8, pp. 2035-2043, Aug 2009. - source
- Not checked — no full text of the sourcecites [85] [84] A. Barachant, S. Bonnet, M. Congedo, and C. Jutten, 'Multiclass Brain-Computer Interface Classification by Riemannian Geometry,' IEEE Transactions on Biomedical Engineering , vol. 59, no. 4, pp. 920-928, Apr. 2012. - source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Could not verify
Every source the paper cites for this claim was checked and supports it.
In the paper
“Here we see the performances of ShallowConvNet and FBCSP are very similar, replicating previous results as reported in [32], while DeepConvNet performance is significantly lower.”
What the sources say
- Supportedcites [32] [32] R. T. Schirrmeister, J. T. Springenberg, L. D. J. Fiederer, M. Glasstetter, K. Eggensperger, M. Tangermann, F. Hutter, W. Burgard, and T. Ball, 'Deep learning with convolutional neural networks for eeg decoding and visualization,' Human Brain Mapping , vol. 38, no. 11, pp. 5391-5420, 2017. [Online]. Available: http://dx.doi.org/10.1002/hbm.23730 - source
Result 2 in the Results section says the shallow ConvNet reached accuracies similar to FBCSP, and Table 2 reports closely comparable mean accuracies across subjects. The citation supports the comparison between ShallowConvNet and FBCSP; the citing authors’ observation about their own DeepConvNet result is not attributed to this source. (LLM, conf=0.93)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“We see that the EEGNet models are up to two orders of magnitude smaller than both DeepConvNet and ShallowConvNet across all datasets.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper states — it cites no source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“but with two orders of magnitude fewer parameters to fit”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
35 more claims
Create a free account to see all 72 claims and to verify the 1 reference still open.
- It may state two separate claims as one. 20
- Probably not a substantial claim. 15
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“It has demonstrated good robustness and has been widely used as a benchmark in EEG analysis”
From A Cascade xDAWN EEGNet Structure for Unified Visual-Evoked Related Potential Detection (2024) · cites this paper as [23]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Notably, the EEGNet exhibits exceptional generalization capabilities, displaying the ability to perform well on diverse datasets, and also exhibits robustness to noise, enabling reliable performance even in the presence of noisy input signals [23].”
From A Cascade xDAWN EEGNet Structure for Unified Visual-Evoked Related Potential Detection (2024) · cites this paper as [23]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“ensuring the reliability of algorithm”
From IENet: a robust convolutional neural network for EEG based brain-computer interfaces (2022) · cites this paper as [13]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“The signal processing of each BCI paradigm requires the participation of relevant prior knowledge [13].”
From IENet: a robust convolutional neural network for EEG based brain-computer interfaces (2022) · cites this paper as [13]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Depthwise and separable convolutions are applied in EEGNet to extract high-dimensional features of EEG, which reduces the number of model parameters to train [13].”
From IENet: a robust convolutional neural network for EEG based brain-computer interfaces (2022) · cites this paper as [13]
Experimental — accuracy not yet measured
17 more claims about this paper
Create a free account to see every claim other papers make about this one.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ 5.1 DeepConvNet and ShallowConvNet architectures · p. 23
“| Conv2D | 200 | (1, 5) | 100 * 200 * C + 200 | Linear | mode = valid, max norm = 2 |”
§ 2.2.1 EEGNet: Compact CNN Architecture · p. 9
“| P300 | 1 | 174,127 | 104,002 | 1,066 | 2,258 |”
Citations
Cites
Very Deep Convolutional Networks for Large-Scale Image Recognition
Simonyan, K., & Zisserman, A. (2014) · arXiv · doi:10.48550/arxiv.1409.1556 · Cited by 114,148
ImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
LeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
Densely Connected Convolutional Networks
Huang, G., Liu, Z., van der Maaten, L., & Weinberger, K. Q. (2016) · arXiv · doi:10.48550/arxiv.1608.06993 · Cited by 48,960
Densely Connected Convolutional Networks
Huang, G., Liu, Z., Van Der Maaten, L., & Weinberger, K. Q. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.243 · Cited by 48,566
Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
Ioffe, S., & Szegedy, C. (2015) · arXiv · doi:10.48550/arxiv.1502.03167 · Cited by 27,818 · cited 1×
“We apply batch normalization [80] along the feature map dimension before applying the exponential linear unit (ELU) nonlinearity [81].”
MethodDelorme, A., & Makeig, S. (2004) · Journal of Neuroscience Methods · doi:10.1016/j.jneumeth.2003.10.009 · Cited by 24,371 · cited 3×
“The data were referenced to linked mastoids, bandpass filtered, using an FIR filter implemented in EEGLAB [53], between 0.”
MethodInfluentialRegularization and Variable Selection Via the Elastic Net
Zou, H., & Hastie, T. (2005) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/j.1467-9868.2005.00503.x · Cited by 20,957
Deep learning in neural networks: an overview
Schmidhuber, J. (2015) · Neural Networks · doi:10.1016/j.neunet.2014.09.003 · Cited by 18,570
"Why Should I Trust You?": Explaining the Predictions of Any Classifier
Ribeiro, M. T., Singh, S., & Guestrin, C. (2016) · Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining · doi:10.1145/2939672.2939778 · Cited by 17,555
Visualizing and understanding convolutional networks
Zeiler, M. D., & Fergus, R. (2014) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-10590-1_53 · Cited by 15,839 · cited 1×
“This last finding is particularly important, as it is a critical component to understanding the validity and robustness of CNN model architectures not just for EEG [32,101], but for CNN architectures in general [16,94,99].”
BackgroundTensorFlow: A system for large-scale machine learning
Abadi, M., Barham, P., Chen, J., Chen, Z., Davis, A., Dean, J., Devin, M., Ghemawat, S., Irving, G., Isard, M., et al. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1605.08695 · Cited by 11,822
The control of the false discovery rate in multiple testing under dependency
Benjamini, Y., & Yekutieli, D. (2001) · The Annals of Statistics · doi:10.1214/aos/1013699998 · Cited by 10,490 · cited 1×
“e reference algorithm. For each EEGNet model conguration a one-way t-test was performed, testing if the model signicantly outperformed the reference algorithm. We use the False Discovery Rate (FDR) [78] procedure to correct for multiple comparisons. We also conduct two additional analyses where model performance is co…”
MethodHinton, G., Deng, L., Yu, D., Dahl, G., Mohamed, A.-R., Jaitly, N., Senior, A., Vanhoucke, V., Nguyen, P., Sainath, T., et al. (2012) · IEEE Signal Processing Magazine · doi:10.1109/msp.2012.2205597 · Cited by 10,260 · cited 1×
“lication domains [3{5,10{12]. Deep Learning has largely alleviated the need for manual feature extraction, achieving state-ofthe-art performance in elds such as computer vision and speech recognition [13,14]. Specically, the use of deep convolutional neural networks (CNNs) has grown due in part to their success in man…”
MethodBrain–computer interfaces for communication and control
Wolpaw, J. R., Birbaumer, N., McFarland, D. J., Pfurtscheller, G., & Vaughan, T. M. (2002) · Clinical Neurophysiology · doi:10.1016/s1388-2457(02)00057-3 · Cited by 7,927 · cited 1×
“A Brain-Computer Interface (BCI) is a mechanism for communicating with a machine via brain signals, bypassing normal neuromuscular outputs [1].”
BackgroundUpdating P300: An integrative theory of P3a and P3b
Polich, J. (2007) · Clinical Neurophysiology · doi:10.1016/j.clinph.2007.04.019 · Cited by 7,919
Event-related EEG/MEG synchronization and desynchronization: basic principles
Pfurtscheller, G., & Lopes da Silva, F. H. (1999) · Clinical Neurophysiology · doi:10.1016/s1388-2457(99)00141-8 · Cited by 7,328 · cited 1×
“1741-2552/18/056013+17$33.00 V J Lawhern et al”
InfluentialFast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)
Clevert, D.-A., Unterthiner, T., & Hochreiter, S. (2015) · arXiv · doi:10.48550/arxiv.1511.07289 · Cited by 4,513 · cited 1×
“lutional Network Architecture Our EEGNet model is shown in Table 2, for EEG trials having Cchannels and Ttime samples. All layers used the Exponential Linear Unit (ELU) non-linear activation function [66]. The model was estimated using Adam, a stochastic optimization algorithm using adaptive moment estimation [67], op…”
MethodDeep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2015) · arXiv · doi:10.48550/arxiv.1512.03385 · Cited by 4,196
Deep learning with convolutional neural networks for EEG decoding and visualization
Schirrmeister, R. T., Springenberg, J. T., Fiederer, L. D. J., Glasstetter, M., Eggensperger, K., Tangermann, M., Hutter, F., Burgard, W., & Ball, T. (2017) · Human Brain Mapping · doi:10.1002/hbm.23730 · Cited by 3,751
Striving for Simplicity: The All Convolutional Net
Springenberg, J. T., Dosovitskiy, A., Brox, T., & Riedmiller, M. (2014) · arXiv · doi:10.48550/arxiv.1412.6806 · Cited by 3,447
Deep Neural Networks are Easily Fooled: High Confidence Predictions for Unrecognizable Images
Nguyen, A., Yosinski, J., & Clune, J. (2014) · arXiv · doi:10.48550/arxiv.1412.1897 · Cited by 3,258
Deep neural networks are easily fooled: High confidence predictions for unrecognizable images
Nguyen, A., Yosinski, J., & Clune, J. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298640 · Cited by 3,258
A well-conditioned estimator for large-dimensional covariance matrices
Ledoit, O., & Wolf, M. (2004) · Journal of Multivariate Analysis · doi:10.1016/s0047-259x(03)00096-4 · Cited by 3,046
Pascual-Marqui, R. D., Michel, C. M., & Lehmann, D. (1994) · International Journal of Psychophysiology · doi:10.1016/0167-8760(84)90014-x · Cited by 2,869
A neural system for error detection and compensation
Gehring, W. J., Goss, B., Coles, M. G. H., Meyer, D. E., & Donchin, E. (1993) · Psychological Science · doi:10.1111/j.1467-9280.1993.tb00586.x · Cited by 2,862
Learning Important Features Through Propagating Activation Differences
Shrikumar, A., Greenside, P., & Kundaje, A. (2017) · arXiv · doi:10.48550/arxiv.1704.02685 · Cited by 2,761
A review of classification algorithms for EEG-based brain–computer interfaces
Lotte, F., Congedo, M., Lécuyer, A., Lamarche, F., & Arnaldi, B. (2007) · Journal of Neural Engineering · doi:10.1088/1741-2560/4/2/r01 · Cited by 2,665
Methods for interpreting and understanding deep neural networks
Montavon, G., Samek, W., & Müller, K.-R. (2018) · Digital Signal Processing · doi:10.1016/j.dsp.2017.10.011 · Cited by 2,665 · cited 1×
“This last finding is particularly important, as it is a critical component to understanding the validity and robustness of CNN model architectures not just for EEG [32,101], but for CNN architectures in general [16,94,99].”
BackgroundBrain Computer Interfaces, a Review
Nicolas-Alonso, L. F., & Gomez-Gil, J. (2012) · Sensors · doi:10.3390/s120201211 · Cited by 2,143 · cited 1×
“processing stages [3]: a data collection stage, where neural data is recorded; a signal processing stage, where the recorded data is preprocessed and cleaned; a feature extraction stage, where meaningful information is extracted from the neural data; a classification stage, where a decision is interpreted from the dat…”
BackgroundMotor imagery and direct brain-computer communication
Pfurtscheller, G., & Neuper, C. (2001) · Proceedings of the IEEE · doi:10.1109/5.939829 · Cited by 1,990 · cited 2×
“In contrast to ERPbased BCIs, which rely mainly on the detection of the ERP waveform from some external event or stimulus, oscillatory BCIs use the signal power of specific EEG frequency bands for external control and are generally asynchronous [47].”
BackgroundMethodFalkenstein, M., Hohnsbein, J., Hoormann, J., & Blanke, L. (1991) · Electroencephalography and Clinical Neurophysiology · doi:10.1016/0013-4694(91)90062-9 · Cited by 1,706 · cited 1×
“They can be observed in a variety of tasks, including time interval production paradigms [50] and in forced-choice paradigms [51, 52].”
BackgroundFeature selection, L 1 vs. L 2 regularization, and rotational invariance
Ng, A. Y. (2004) · Twenty-first international conference on Machine learning - ICML '04 · doi:10.1145/1015330.1015435 · Cited by 1,572
Miltner, W. H. R., Braun, C. H., & Coles, M. G. H. (1997) · Journal of Cognitive Neuroscience · doi:10.1162/jocn.1997.9.6.788 · Cited by 1,530
Going Deeper with Convolutions
Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., & Rabinovich, A. (2014) · arXiv · doi:10.48550/arxiv.1409.4842 · Cited by 1,461 · cited 1×
“This parametrization trick has been used by several groups [16,17] as a way to efficiently build deep convolutional networks for image analysis.”
MethodThe PREP pipeline: standardized preprocessing for large-scale EEG analysis
Bigdely-Shamlo, N., Mullen, T., Kothe, C., Su, K.-M., & Robbins, K. A. (2015) · Frontiers in Neuroinformatics · doi:10.3389/fninf.2015.00016 · Cited by 1,351 · cited 1×
“Due to extensive signal noise present in the data, the EEG data were first processed with the PREP pipeline [61].”
MethodReview of the BCI Competition IV
Tangermann, M., Müller, K.-R., Aertsen, A., Birbaumer, N., Braun, C., Brunner, C., Leeb, R., Mehring, C., Miller, K. J., Müller-Putz, G. R., et al. (2012) · Frontiers in Neuroscience · doi:10.3389/fnins.2012.00055 · Cited by 1,288 · cited 1×
“The EEG data used here was Dataset 2A from BCI Competition IV [62] (called the SMR dataset for the remainder of the manuscript).”
MethodFilter Bank Common Spatial Pattern Algorithm on BCI Competition IV Datasets 2a and 2b
Ang, K. K., Chin, Z. Y., Wang, C., Guan, C., & Zhang, H. (2012) · Frontiers in Neuroscience · doi:10.3389/fnins.2012.00039 · Cited by 1,264
Event-related cortical desynchronization detected by power measurements of scalp EEG
Pfurtscheller, G., & Aranibar, A. (1977) · Electroencephalography and Clinical Neurophysiology · doi:10.1016/0013-4694(77)90235-8 · Cited by 1,108
Bashashati, A., Fatourechi, M., Ward, R. K., & Birch, G. E. (2007) · Journal of Neural Engineering · doi:10.1088/1741-2560/4/2/r03 · Cited by 907 · cited 5×
“For example, the [16, 4] × [2, 16] model configuration, while performing the best on P300, was slightly worse in SMR and among the worst models for MRCP.”
BackgroundMethodInfluentialZander, T. O., & Kothe, C. (2011) · Journal of Neural Engineering · doi:10.1088/1741-2560/8/2/025005 · Cited by 887
Multiclass Brain–Computer Interface Classification by Riemannian Geometry
Barachant, A., Bonnet, S., Congedo, M., & Jutten, C. (2012) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2011.2172210 · Cited by 878 · cited 2×
“Project the covariance matrices onto the tangent space using the log-Euclidean metric [85, 89].”
MethodCombining brain-computer interfaces and assistive technologies: state-of-the-art and challenges
Millán, J. D. R. (2010) · Frontiers in Neuroscience · doi:10.3389/fnins.2010.00161 · Cited by 866
A novel deep learning approach for classification of EEG motor imagery signals
Tabar, Y. R., & Halici, U. (2017) · Journal of Neural Engineering · doi:10.1088/1741-2560/14/1/016003 · Cited by 859 · cited 2×
“For example, the model proposed in [36] uses the time-frequency decomposition of channels Cz, C3 and C4 as the inputs for a motor imagery classification task.”
MethodAuditory event-related dynamics of the EEG spectrum and effects of exposure to tones
Makeig, S. (1993) · Electroencephalography and Clinical Neurophysiology · doi:10.1016/0013-4694(93)90110-h · Cited by 844
Learning Representations from EEG with Deep Recurrent-Convolutional Neural Networks
Bashivan, P., Rish, I., Yeasin, M., & Codella, N. (2015) · arXiv · doi:10.48550/arxiv.1511.06448 · Cited by 836 · cited 3×
“Recently, a recurrent convolutional network was proposed by [28] to predict mental workload.”
BackgroundMethodBrain-Controlled Interfaces: Movement Restoration with Neural Prosthetics
Schwartz, A. B., Cui, X. T., Weber, D. J., & Moran, D. W. (2006) · Neuron · doi:10.1016/j.neuron.2006.09.019 · Cited by 792
Convolutional Neural Networks for P300 Detection with Application to Brain-Computer Interfaces
Cecotti, H., & Graser, A. (2011) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2010.125 · Cited by 778 · cited 2×
“quency-transformed EEG signals for classication among dierent classes of listened music [34,38]. Temporal convolutional networks have also been used for the detection of visual-evoked potentials in [39] and in [33]. However, a comprehensive analysis of the performance of a single model architecture across multiple EEG…”
MethodToward enhanced p300 speller performance
Krusienski, D. J., Sellers, E. W., McFarland, D. J., Vaughan, T. M., & Wolpaw, J. R. (2008) · Journal of Neuroscience Methods · doi:10.1016/j.jneumeth.2007.07.017 · Cited by 749
Deecke, L., Scheid, P., & Kornhuber, H. (1969) · Experimental Brain Research · doi:10.1007/bf00235441 · Cited by 748 · cited 1×
“The MRCP components can be seen before movement onset (a slow 0–5 Hz readiness potential [66, 67] and an early desynchronization in the 10–12 Hz frequency band), at movement onset (a slow motor potential [67, 68]), and after movement onset (a late synchronization of 20–30 Hz activity approximately 1 s after movement e…”
BackgroundXception: Deep Learning with Depthwise Separable Convolutions
Chollet, F. (2016) · arXiv · doi:10.48550/arxiv.1610.02357 · Cited by 538
xDAWN Algorithm to Enhance Evoked Potentials: Application to Brain–Computer Interface
Rivet, B., Souloumiac, A., Attina, V., & Gibert, G. (2009) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2009.2012869 · Cited by 537 · cited 1×
“ach is the approach which won the Kaggle BCI Competition (code and documenta9 tion at http://github.com/alexandrebarachant/bci-challenge-ner-2015), which uses a combination of xDAWN Spatial Filtering [83], Riemannian Geometry [84,85], channel subset selection and L 1 feature regularization (referred to as xDAWN + RG f…”
MethodVisualizing and Understanding Convolutional Networks
Zeiler, M. D., & Fergus, R. (2013) · arXiv · doi:10.48550/arxiv.1311.2901 · Cited by 522
Classification of covariance matrices using a Riemannian-based kernel for BCI applications
Barachant, A., Bonnet, S., Congedo, M., & Jutten, C. (2013) · Neurocomputing · doi:10.1016/j.neucom.2012.12.039 · Cited by 493 · cited 1×
“Project the covariance matrices onto the tangent space using the log-euclidean metric [84,88].”
MethodA Deep Learning Scheme for Motor Imagery Classification based on Restricted Boltzmann Machines
Lu, N., Li, T., Ren, X., & Miao, H. (2017) · IEEE Transactions on Neural Systems and Rehabilitation Engineering · doi:10.1109/tnsre.2016.2601240 · Cited by 419 · cited 1×
“Restricted Boltzman machines (RBMs) have been used for motor imagery [41].”
MethodBCILAB: a platform for brain–computer interface development
Kothe, C. A., & Makeig, S. (2013) · Journal of Neural Engineering · doi:10.1088/1741-2560/10/5/056014 · Cited by 412
Interpretable Deep Neural Networks for Single-Trial EEG Classification
Sturm, I., Lapuschkin, S., Samek, W., & Müller, K.-R. (2016) · Journal of Neuroscience Methods · doi:10.48550/arxiv.1604.08201 · Cited by 407
Classification of patterns of EEG synchronization for seizure prediction
Mirowski, P., Madhavan, D., LeCun, Y., & Kuzniecky, R. (2009) · Clinical Neurophysiology · doi:10.1016/j.clinph.2009.09.002 · Cited by 380
Interpretable deep neural networks for single-trial EEG classification
Sturm, I., Lapuschkin, S., Samek, W., & Müller, K.-R. (2016) · Journal of Neuroscience Methods · doi:10.1016/j.jneumeth.2016.10.008 · Cited by 374 · cited 3×
“For example, CNNs have been used for epilepsy prediction and monitoring [21–25], for auditory music retrieval [26, 27], for detection of visual-evoked responses [28–31] and for motor imagery classification [32, 33], while deep belief networks (DBNs) have been used for sleep stage detection [34], anomaly detection [35]…”
BackgroundMethodTowards better understanding of gradient-based attribution methods for Deep Neural Networks
Ancona, M., Ceolini, E., Öztireli, C., & Groß, M. H. (2017) · Zurich Open Repository and Archive (University of Zurich) · doi:10.3929/ethz-b-000249929 · Cited by 368
Brain-Computer Interfaces: Beyond Medical Applications
van Erp, J., Lotte, F., & Tangermann, M. (2012) · Computer · doi:10.1109/mc.2012.107 · Cited by 362
P300 brain computer interface: current challenges and emerging trends
Fazel-Rezai, R., Allison, B. Z., Guger, C., Sellers, E. W., Kleih, S. C., & Kübler, A. (2012) · Frontiers in Neuroengineering · doi:10.3389/fneng.2012.00014 · Cited by 359 · cited 1×
“They are generally robust across subjects and contain well-stereotyped waveforms, enabling the exact time course of the ERP to be modeled through machine learning efficiently [40].”
BackgroundThe Berlin Brain–Computer Interface: Non-Medical Uses of BCI Technology
Blankertz, B., Tangermann, M., Vidaurre, C., Fazli, S., Sannelli, C., Haufe, S., Maeder, C., Ramsey, L., Sturm, I., Curio, G., et al. (2010) · Frontiers in Neuroscience · doi:10.3389/fnins.2010.00198 · Cited by 357
Lotte, F. (2015) · Proceedings of the IEEE · doi:10.1109/jproc.2015.2404941 · Cited by 327
Event-related desynchronization and movement-related cortical potentials on the ECoG and EEG
Toro, C., Deuschl, G., Thatcher, R., Sato, S., Kufta, C., & Hallett, M. (1994) · Electroencephalography and Clinical Neurophysiology/Evoked Potentials Section · doi:10.1016/0168-5597(94)90126-0 · Cited by 316 · cited 1×
“One particular example of this is the movement-related cortical potential (MRCP), which can be elicited by voluntary movements of the hands and feet and is observable through EEG along the central and midline electrodes, contralateral to the hand or foot movement [55].”
BackgroundEvent-related desynchronization and movement-related cortical potentials on the ECoG and EEG
TORO, C., DEUSCHL, G., THATCHER, R., SATO, S., KUFTA, C., & HALLETT, M. (1994) · Electroencephalography and Clinical Neurophysiology · doi:10.1016/0013-4694(94)00177-m · Cited by 255
Sleep Stage Classification Using Unsupervised Feature Learning
Längkvist, M., Karlsson, L., & Loutfi, A. (2012) · Advances in Artificial Neural Systems · doi:10.1155/2012/107046 · Cited by 248 · cited 1×
“For example, L¨angkvist et. al. used deep belief networks (DBNs) and restricted boltzmann machines (RBMs), together with hidden Markov models, to temporally model different stages of sleep using EEG [36].”
BackgroundBrain–Computer Interface Technologies in the Coming Decades
Lance, B. J., Kerick, S. E., Ries, A. J., Oie, K. S., & McDowell, K. (2012) · Proceedings of the IEEE · doi:10.1109/jproc.2012.2184830 · Cited by 231 · cited 3×
“The EEGNet model with the [2, 32] × [8, 4] configuration will be used for the remainder of the manuscript.”
BackgroundMethodSelf-citationYang, Z., Moczulski, M., Denil, M., De Freitas, N., Song, L., & Wang, Z. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.173 · Cited by 215 · cited 1×
“We also show that EEGNet performs just as well as a more paradigm-specific EEG CNN model, but with two orders of magnitude fewer parameters to fit, representing a more efficient use of model parameters (an aspect that has been explored in previous deep learning literature, see [42,43]).”
MethodBCI meeting 2005-workshop on BCI signal processing: feature extraction and translation
McFarland, D. J., Anderson, C. W., Muller, K.-R., Schlogl, A., & Krusienski, D. J. (2006) · IEEE Transactions on Neural Systems and Rehabilitation Engineering · doi:10.1109/tnsre.2006.875637 · Cited by 204
Wulsin, D. F., Gupta, J. R., Mani, R., Blanco, J. A., & Litt, B. (2011) · Journal of Neural Engineering · doi:10.1088/1741-2560/8/3/036015 · Cited by 185 · cited 1×
“This raises the following question: Can Deep Learning approaches be used to design better features suitable for classification of EEG signals? While there have been many previous works on applying Deep Learning approaches to EEG signals (for example, epilepsy prediction and monitoring [21–25], sleep stage detection [2…”
MethodA Deep Learning Method for Classification of EEG Data Based on Motor Imagery
An, X., Kuang, D., Guo, X., Zhao, Y., & He, L. (2014) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-09330-7_25 · Cited by 183
Decoding Individual Finger Movements from One Hand Using Human EEG Signals
Liao, K., Xiao, R., Gonzalez, J., & Ding, L. (2014) · PLoS ONE · doi:10.1371/journal.pone.0085192 · Cited by 174 · cited 1×
“This produced the well-known alpha and beta synchronizations (i.e. increases in power) and desynchronizations (i.e. decreases in power), most clearly observed over the contralateral motor cortex [60–62].”
BackgroundCross-session classification of mental workload levels using EEG and an adaptive deep learning model
Yin, Z., & Zhang, J. (2017) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2016.11.013 · Cited by 163 · cited 1×
“An adaptive method based on stacked denoising autoencoders has been proposed for mental workload classification [41]).”
MethodLearning Robust Features using Deep Learning for Automatic Seizure Detection
Thodoroff, P., Pineau, J., & Lim, A. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1608.00220 · Cited by 163
EEG spectral dynamics during discrimination of auditory and visual targets
Mazaheri, A., & Picton, T. W. (2005) · Cognitive Brain Research · doi:10.1016/j.cogbrainres.2004.12.013 · Cited by 162 · cited 1×
“appears to show a significant theta-beta relationship; while increases in theta activity have been previously noted in the P300 literature in response to targets [102], a relationship between theta and beta has not previously been noted.”
BackgroundCortical potentials preceding voluntary movement: Evidence for three periods of preparation in man
Barrett, G., Shibasaki, H., & Neshige, R. (1986) · Electroencephalography and Clinical Neurophysiology · doi:10.1016/0013-4694(86)90017-9 · Cited by 156 · cited 1×
“The MRCP components can be seen before movement onset (a slow 0-5Hz readiness potential [65, 66] and an early desynchronization in the 10-12Hz frequency band), at movement onset (a slow motor potential [66,67]), and after movement onset (a late synchronization of 20-30Hz activity approximately 1 second after movement…”
BackgroundIn a Blink of an Eye and a Switch of a Transistor: Cortically Coupled Computer Vision
Sajda, P., Pohlmeyer, E., Wang, J., Parra, L. C., Christoforou, C., Dmochowski, J., Hanna, B., Bahlmann, C., Singh, M. K., & Chang, S.-F. (2010) · Proceedings of the IEEE · doi:10.1109/jproc.2009.2038406 · Cited by 153 · cited 1×
“When the image presentation rate increases to 2 Hz or more, it is commonly referred to as rapid serial visual presentation (RSVP), which has been used in several BCIs for large image database triage [43,48,49]. The EEG data used here has been previously described in [49]; a brief description is given below. 18 partici…”
MethodObjective and Subjective Evaluation of Online Error Correction during P300-Based Spelling
Margaux, P., Emmanuel, M., Sébastien, D., Olivier, B., & Jérémie, M. (2012) · Advances in Human-Computer Interaction · doi:10.1155/2012/578295 · Cited by 149 · cited 2×
“The feedback ERN is characterized as a large negative deflection approximately 300 ms after feedback, followed by a positive component 500 ms to 1 s after feedback (see Figure 7 of [50] for an illustration).”
BackgroundMethodLawhern, V. J., Hairston, W. D., McDowell, K., Westerfield, M., & Robbins, K. A. (2012) · Journal of Neuroscience Methods · doi:10.1016/j.jneumeth.2012.05.017 · Cited by 145
Self-citationSpüler, M., Bensch, M., Kleih, S., Rosenstiel, W., Bogdan, M., & Kübler, A. (2012) · Clinical Neurophysiology · doi:10.1016/j.clinph.2011.11.082 · Cited by 144
Convolutional Neural Network for Multi-Category Rapid Serial Visual Presentation BCI
Manor, R., & Geva, A. B. (2015) · Frontiers in Computational Neuroscience · doi:10.3389/fncom.2015.00146 · Cited by 144 · cited 6×
“For all ERP-based analyses we compare EEGNet to the convolutional network model by Manor and Geva [32] (see Table 1 of [32]).”
BackgroundMethodInfluentialDeep learning for epileptic intracranial EEG data
Antoniades, A., Spyrou, L., Took, C. C., & Sanei, S. (2016) · 2016 IEEE 26th International Workshop on Machine Learning for Signal Processing (MLSP) · doi:10.1109/mlsp.2016.7738824 · Cited by 129
The Emerging World Of Motor Neuroprosthetics: a Neurosurgical Perspective
Leuthardt, E. C., Schalk, G., Moran, D., & Ojemann, J. G. (2006) · Neurosurgery · doi:10.1227/01.neu.0000221506.06947.ac · Cited by 128
Parallel convolutional-linear neural network for motor imagery classification
Sakhavi, S., Guan, C., & Yan, S. (2015) · 2015 23rd European Signal Processing Conference (EUSIPCO) · doi:10.1109/eusipco.2015.7362882 · Cited by 128 · cited 1×
“5] seconds post cue onset (the same window which was used in [38,39]).”
BackgroundCecotti, H., Eckstein, M. P., & Giesbrecht, B. (2014) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2014.2302898 · Cited by 109 · cited 1×
“We believe models that fall in (1), such as EEGNet and others [31, 33, 34], offer the best tradeoff between input dimensionality and the flexibility to discover relevant features by providing all available channels.”
MethodA Plug&Play P300 BCI Using Information Geometry
Barachant, A., & Congedo, M. (2014) · arXiv · doi:10.48550/arxiv.1409.0107 · Cited by 102
The extraction of motion-onset VEP BCI features based on deep learning and compressed sensing
Ma, T., Li, Y., Yang, H., Lv, X., Li, P., Liu, T., Yao, D., & Xu, P. (2016) · Journal of Neuroscience Methods · doi:10.1016/j.jneumeth.2016.11.002 · Cited by 82 · cited 1×
“For example, CNNs have been used for epilepsy prediction and monitoring [21–25], for auditory music retrieval [26, 27], for detection of visual-evoked responses [28–31] and for motor imagery classification [32, 33], while deep belief networks (DBNs) have been used for sleep stage detection [34], anomaly detection [35]…”
MethodZander, T. O., Kothe, C. A., Welke, S., & Roetting, M. (2009) · Lecture notes in computer science · doi:10.1007/978-3-642-02812-0_86 · Cited by 80
Waytowich, N. R., Lawhern, V. J., Bohannon, A. W., Ball, K. R., & Lance, B. J. (2016) · Frontiers in Neuroscience · doi:10.3389/fnins.2016.00430 · Cited by 79 · cited 1×
“While within-subject models tend to perform better than cross-subject models on a variety of tasks, there is ongoing research investigating techniques to minimize (or possibly eliminate) the need for subject-specific information to train robust systems [45, 52].”
BackgroundSelf-citationBilinear Discriminant Component Analysis
Dyrholm, M., Christoforou, C., & Parra, L. (2007) · Journal of machine learning research · doi:10.5555/1314498.1314535 · Cited by 76
Wearable seizure detection using convolutional neural networks with transfer learning
Page, A., Shea, C., & Mohsenin, T. (2016) · 2016 IEEE International Symposium on Circuits and Systems (ISCAS) · doi:10.1109/iscas.2016.7527433 · Cited by 70
A New Generation of Brain-Computer Interface Based on Riemannian Geometry
Congedo, M., Barachant, A., & Andreev, A. (2013) · arXiv · doi:10.48550/arxiv.1310.8115 · Cited by 69 · cited 1×
“f the approach, which is done in ve steps: 1.Train two set of 5 xDAWN spatial lters, one set for each class of a binary classication task, using the ERP template concatenation method as described in [85,86]. 2.Perform EEG electrode selection through backward elimination [87] to keep only the most relevant 35 channels.…”
MethodChannel selection procedure using riemannian distance for BCI applications
Barachant, A., & Bonnet, S. (2011) · International IEEE/EMBS Conference on Neural Engineering · doi:10.1109/ner.2011.5910558 · Cited by 66 · cited 1×
“atial lters, one set for each class of a binary classication task, using the ERP template concatenation method as described in [85,86]. 2.Perform EEG electrode selection through backward elimination [87] to keep only the most relevant 35 channels. 3.Project the covariance matrices onto the tangent space using the log-…”
MethodInfluentialMarathe, A. R., Lawhern, V. J., Wu, D., Slayback, D., & Lance, B. J. (2016) · IEEE Transactions on Neural Systems and Rehabilitation Engineering · doi:10.1109/tnsre.2015.2502323 · Cited by 64 · cited 1×
“The EEG data used here were previously described in [45]; a brief description is given below.”
MethodSelf-citationKarimi, F., Kofman, J., Mrachacz‐Kersting, N., Farina, D., & Jiang, N. (2017) · Frontiers in Neuroscience · doi:10.3389/fnins.2017.00356 · Cited by 61
Deep Feature Learning for EEG Recordings
Stober, S., Sternin, A., Owen, A. M., & Grahn, J. A. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1511.04306 · Cited by 61
Yom‐Tov, E., & Inbar, G. F. (2003) · Medical & Biological Engineering & Computing · doi:10.1007/bf02343543 · Cited by 54
How to Explain Individual Classification Decisions
(2010) · Journal of Machine Learning Research · doi:10.5555/1756006.1859912 · Cited by 54
Movement related slow cortical potentials in severely paralyzed chronic stroke patients
Yilmaz, Ö., Birbaumer, N., & Ramos‐Murguialday, A. (2015) · Frontiers in Human Neuroscience · doi:10.3389/fnhum.2014.01033 · Cited by 49 · cited 1×
“The MRCP components can be seen before movement onset (a slow 0-5Hz readiness potential [65, 66] and an early desynchronization in the 10-12Hz frequency band), at movement onset (a slow motor potential [66,67]), and after movement onset (a late synchronization of 20-30Hz activity approximately 1 second after movement…”
BackgroundPredicting Seizures from Electroencephalography Recordings: A Knowledge Transfer Strategy
Liang, J., Lu, R., Zhang, C., & Wang, F. (2016) · 2016 IEEE International Conference on Healthcare Informatics (ICHI) · doi:10.1109/ichi.2016.27 · Cited by 49
Single-trial EEG RSVP classification using convolutional neural networks
Shamwell, J., Lee, H., Kwon, H., Marathe, A. R., Lawhern, V., & Nothwang, W. (2016) · SPIE Proceedings · doi:10.1117/12.2224172 · Cited by 34 · cited 1×
“We believe models that fall in (1), such as EEGNet and others [28, 30, 31], offer the best tradeoff between input dimensionality and the flexibility to discover relevant features by providing all available channels.”
MethodSelf-citationAuditory Event-Related Dynamics of the EEG Spectrum and Effects of Exposure to Tones
Makeig, S. (1992) · PsycEXTRA Dataset · doi:10.1037/e640342009-001 · Cited by 32
Huang, D.-S., Han, K., & Gromiha, M. M. (2014) · Lecture notes in computer science · doi:10.1007/978-3-319-09330-7 · Cited by 24
Informed decomposition of electroencephalographic data
Gordon, S. M., Lawhern, V. J., Passaro, A. D., & McDowell, K. (2015) · Journal of Neuroscience Methods · doi:10.1016/j.jneumeth.2015.08.019 · Cited by 23 · cited 1×
“The MRCP has been used previously to develop motor control BCIs for both healthy and physically disabled patients [69–71] The EEG data used here have been previously described in [72]; a brief description is given below.”
MethodSelf-citationReal World BCI: Cross-Domain Learning and Practical Applications
Gordon, S. M., Jaswa, M., Solon, A. J., & Lawhern, V. J. (2017) · Proceedings of the 2017 ACM Workshop on An Application-oriented Approach to BCI out of the laboratory · doi:10.1145/3038439.3038444 · Cited by 20
Self-citationCortically Coupled Computing: A New Paradigm for Synergistic Human-Machine Interaction
Saproo, S., Faller, J., Shih, V., Sajda, P., Waytowich, N. R., Bohannon, A., Lawhern, V. J., Lance, B. J., & Jangraw, D. (2016) · Computer · doi:10.1109/mc.2016.294 · Cited by 18
Self-citationThe Challenges of Using Scalp-EEG Input Signals for Continuous Device Control
Johnson, G. D., Waytowich, N. R., & Krusienski, D. J. (2011) · Lecture notes in computer science · doi:10.1007/978-3-642-21852-1_60 · Cited by 6 · cited 1×
“This is especially true when CNNs are used for the analysis of EEG data where features from neural signals are often non-stationary and corrupted by noise artifacts [103, 104].”
BackgroundSelf-citationTHE EMERGING WORLD OF MOTOR NEUROPROSTHETICS: A NEUROSURGICAL PERSPECTIVE
Leuthardt, E. C., Schalk, G., Moran, D. W., & Ojemann, J. G. (2006) · Neurosurgery · doi:10.1227/01.neu.0000243275.01470.c0 · Cited by 6
Deep Learning in Neural Networks: An Overview
Schmidhuber, J. (2014) · arXiv · doi:10.48550/arxiv.1404.7828
Carretero (1998) · doi:10.1175/1520-0477(1998)079<0741:cwasit>2.0.co;2
Untitled work
Srivastava (2014) · doi:10.5555/2627435.2670313
Research works printed but not linked: 1
- [26] S. Stober, D. J. Cameron, and J. A. Grahn, 'Using convolutional neural networks to recognize rhythm stimuli from electroencephalography recordings,' in Advances in Neural Information Processing Systems 27 , Z. Ghahramani, M. Welling, C. Cortes, N. D. Lawrence, and K. Q. Weinberger, Eds. Curran Associates, Inc., 2014, pp. 1449-1457. -
Cited by
Deep learning with convolutional neural networks for EEG decoding and visualization
Schirrmeister, R. T., Springenberg, J. T., Fiederer, L. D. J., Glasstetter, M., Eggensperger, K., Tangermann, M., Hutter, F., Burgard, W., & Ball, T. (2017) · Human Brain Mapping · doi:10.1002/hbm.23730 · Cited by 3,751
A review of classification algorithms for EEG-based brain–computer interfaces: A 10 year update
Lotte, F., Bougrain, L., Cichocki, A., Clerc, M., Congedo, M., Rakotomamonjy, A., & Yger, F. (2018) · Journal of Neural Engineering · doi:10.1088/1741-2552/aab2f2 · Cited by 2,239
Deep learning for electroencephalogram (EEG) classification tasks: a review
Craik, A., He, Y., & Contreras-Vidal, J. L. (2019) · Journal of Neural Engineering · doi:10.1088/1741-2552/ab0ab5 · Cited by 1,758
“1 FC 20 h 2 OUT [94] CNN ELU (conv) SV P300(1) N/A P300(1) P300(1) 3 conv Softmax (FC) All (64/56/22) 1–40 Hz P300(2) 1–40 Hz MI 4–40 Hz 15 93% 1 FC 0.”
Deep learning-based electroencephalography analysis: a systematic review
Roy, Y., Banville, H., Albuquerque, I., Gramfort, A., Falk, T. H., & Faubert, J. (2019) · Journal of Neural Engineering · doi:10.1088/1741-2552/ab260c · Cited by 1,532
“The analysis of model activations was used in multiple studies [93, 97, 120, 172, 186, 214, 227, 231].”
MethodEEG Conformer: Convolutional Transformer for EEG Decoding and Visualization
Song, Y., Zheng, Q., Liu, B., & Gao, X. (2023) · IEEE Transactions on Neural Systems and Rehabilitation Engineering · doi:10.1109/tnsre.2022.3230250 · Cited by 1,095
“We use exponential linear units (ELUs) as the activation function for nonlinearity following [17].”
BackgroundMethodInfluential
Metadata
- Authors
- Lawhern, V. J., Solon, A. J., Waytowich, N. R., Gordon, S. M., Hung, C. P., & Lance, B. J.
- Affiliations
- DEVCOM Army Research Laboratory (US)
- Corresponding author
- Lawhern, Vernon J
- Venue
- Journal of Neural Engineering
- Publisher
- IOP Publishing
- Published
- 2018-07-27
- Volume
- 15
- Issue
- 5
- Pages
- 056013
- ISSN
- 1741-2552 (ISSN-L)
- 1741-2560
- OpenAlex
- W2559463885
- Cited by
- 5,115
- References
- 78
- Citations by year
- 2026: 1132 · 2025: 1233 · 2024: 935 · 2023: 649 · 2022: 468 · 2021: 334 · 2020: 212 · 2019: 105 · 2018: 37
- FWCI
- 102.30
- Type
- journal article
- Language
- en
- Open access
- bronze
- doi.org
- Licence
- iopscience.iop.org/info/page/text-and-data-mining (tdm)
- iopscience.iop.org/page/copyright (vor)
- Locations
- Journal of Neural Engineering (publishedVersion)
- PubMed (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- U.S. Army Research Laboratory (CAST 076910227001, ARL-74A-HRCYB, ARL-H70-HR52, W911NF-10-2-0022)
- Topics
- Life Sciences › Neuroscience › Cognitive Neuroscience › EEG and Brain-Computer Interfaces
- Physical Sciences › Computer Science › Artificial Intelligence › Emotion and Mood Recognition
- Physical Sciences › Computer Science › Human-Computer Interaction › Gaze Tracking and Assistive Technology
- Keywords
- EEGNet, EEG-based brain-computer interface, compact convolutional neural networks, depthwise separable convolution, error-related negativity, movement-related cortical potentials, sensory motor rhythm, EEG classification, model interpretability, feature visualization, limited training data
- MeSH terms
- Adolescent, Adult, Algorithms, Electroencephalography, Evoked Potentials, Visual, Female, Humans, Male, Middle Aged, Movement, Neural Networks, Computer, Event-Related Potentials, P300, Young Adult, Brain-Computer Interfaces
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (qwen3.8-omni-flash and gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.