EEGNet: A Compact Convolutional Network for EEG-based Brain-Computer Interfaces

Abstract

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, 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. In this work we introduce EEGNet, a compact convolutional 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 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). We show that EEGNet generalizes across paradigms better than the reference algorithms when only limited training data is available. We demonstrate three different approaches to visualize the contents of a trained EEGNet model to enable interpretation of the learned features. Our results suggest that EEGNet is robust enough to learn a wide variety of interpretable features over a range of BCI tasks, suggesting that the observed performances were not due to artifact or noise sources in the data.

30 pages, 10 figures. Added additional feature relevance analyses. Minor change to EEGNet architecture. Source code can be found at https://github.com/vlawhern/arl-eegmodels

References in corpus (3)

• Deep learning with convolutional neural networks for EEG decoding and visualization
• Methods for Interpreting and Understanding Deep Neural Networks
• Brain Computer Interface Technologies in the Coming Decades

Cited by in corpus (119)

• Deep learning with convolutional neural networks for EEG decoding and visualization
• A deep learning architecture for temporal sleep stage classification using multivariate and multimodal time series
• EEG based Emotion Recognition: A Tutorial and Review
• TSception: Capturing Temporal Dynamics and Spatial Asymmetry from EEG for Emotion Recognition
• Contrastive Learning of Subject-Invariant EEG Representations for Cross-Subject Emotion Recognition
• Compact Convolutional Neural Networks for Classification of Asynchronous Steady-state Visual Evoked Potentials
• Machine-Learning-Based Diagnostics of EEG Pathology
• Decoding speech perception from non-invasive brain recordings
• MIN2Net: End-to-End Multi-Task Learning for Subject-Independent Motor Imagery EEG Classification
• GCNs-Net: A Graph Convolutional Neural Network Approach for Decoding Time-resolved EEG Motor Imagery Signals
• EEG-based Cross-Subject Driver Drowsiness Recognition with an Interpretable Convolutional Neural Network
• EEG-ITNet: An Explainable Inception Temporal Convolutional Network for Motor Imagery Classification
• A Compact and Interpretable Convolutional Neural Network for Cross-Subject Driver Drowsiness Detection from Single-Channel EEG
• PGCN: Pyramidal Graph Convolutional Network for EEG Emotion Recognition
• Transfer Learning for Motor Imagery Based Brain-Computer Interfaces: A Complete Pipeline
• Deep Learning in EEG: Advance of the Last Ten-Year Critical Period
• EEGFuseNet: Hybrid Unsupervised Deep Feature Characterization and Fusion for High-Dimensional EEG with An Application to Emotion Recognition
• An Accurate EEGNet-based Motor-Imagery Brain-Computer Interface for Low-Power Edge Computing
• Federated Transfer Learning for EEG Signal Classification
• Adversarial Deep Learning in EEG Biometrics
• Tensor-CSPNet: A Novel Geometric Deep Learning Framework for Motor Imagery Classification
• A Decomposition-Based Hybrid Ensemble CNN Framework for Driver Fatigue Recognition
• Efficacy of Transformer Networks for Classification of Raw EEG Data
• CarePre: An Intelligent Clinical Decision Assistance System
• Edge Deep Learning for Neural Implants
• Graph Neural Networks on SPD Manifolds for Motor Imagery Classification: A Perspective from the Time-Frequency Analysis
• Fuzzy temporal convolutional neural networks in P300-based Brain-computer interface for smart home interaction
• Decoding ECoG signal into 3D hand translation using deep learning
• Automated Detection of Abnormalities from an EEG Recording of Epilepsy Patients With a Compact Convolutional Neural Network
• Spatio-Temporal EEG Representation Learning on Riemannian Manifold and Euclidean Space
• Deep learning-based classification of fine hand movements from low frequency EEG
• Subject-Independent Drowsiness Recognition from Single-Channel EEG with an Interpretable CNN-LSTM model
• Ballistocardiogram artifact reduction in simultaneous EEG-fMRI using deep learning
• CSP-Net: Common Spatial Pattern Empowered Neural Networks for EEG-Based Motor Imagery Classification
• Deep comparisons of Neural Networks from the EEGNet family
• Bridging the Gap Between Patient-specific and Patient-independent Seizure Prediction via Knowledge Distillation
• Adversarial Attacks and Defenses in Physiological Computing: A Systematic Review
• Data Augmentation for Seizure Prediction with Generative Diffusion Model
• Multi-Branch Mutual-Distillation Transformer for EEG-Based Seizure Subtype Classification
• Towards Best Practice of Interpreting Deep Learning Models for EEG-based Brain Computer Interfaces
• Evidence of Task-Independent Person-Specific Signatures in EEG using Subspace Techniques
• Channel Reflection: Knowledge-Driven Data Augmentation for EEG-Based Brain-Computer Interfaces
• T-TIME: Test-Time Information Maximization Ensemble for Plug-and-Play BCIs
• Decoding Imagined Speech and Computer Control using Brain Waves
• Federated Motor Imagery Classification for Privacy-Preserving Brain-Computer Interfaces
• EEG Decoding for Datasets with Heterogenous Electrode Configurations using Transfer Learning Graph Neural Networks
• Clinical Brain-Computer Interface Challenge 2020 (CBCIC at WCCI2020): Overview, methods and results
• MixNet: Joining Force of Classical and Modern Approaches Toward the Comprehensive Pipeline in Motor Imagery EEG Classification
• User Identity Protection in EEG-based Brain-Computer Interfaces
• Acting Thoughts: Towards a Mobile Robotic Service Assistant for Users with Limited Communication Skills
• Motor Imagery Classification for Asynchronous EEG-Based Brain-Computer Interfaces
• A large-scale evaluation framework for EEG deep learning architectures
• Towards Understanding Human Functional Brain Development with Explainable Artificial Intelligence: Challenges and Perspectives
• Amplifying Pathological Detection in EEG Signaling Pathways through Cross-Dataset Transfer Learning
• Alignment-Based Adversarial Training (ABAT) for Improving the Robustness and Accuracy of EEG-Based BCIs
• Canine EEG Helps Human: Cross-Species and Cross-Modality Epileptic Seizure Detection via Multi-Space Alignment
• A Generalization of the Sugeno integral to aggregate Interval-valued data: an application to Brain Computer Interface and Social Network Analysis
• Adversarial Filtering Based Evasion and Backdoor Attacks to EEG-Based Brain-Computer Interfaces
• Revisiting Euclidean Alignment for Transfer Learning in EEG-Based Brain-Computer Interfaces
• EEG-based Texture Roughness Classification in Active Tactile Exploration with Invariant Representation Learning Networks
• Interval-valued aggregation functions based on moderate deviations applied to Motor-Imagery-Based Brain Computer Interface
• On-device Learning of EEGNet-based Network For Wearable Motor Imagery Brain-Computer Interface
• EEG Right & Left Voluntary Hand Movement-based Virtual Brain-Computer Interfacing Keyboard Using Hybrid Deep Learning Approach
• Active Poisoning: Efficient Backdoor Attacks on Transfer Learning-Based Brain-Computer Interfaces
• DrowzEE-G-Mamba: Leveraging EEG and State Space Models for Driver Drowsiness Detection
• Spatial Filtering Pipeline Evaluation of Cortically Coupled Computer Vision System for Rapid Serial Visual Presentation
• Enhancing Eye-Tracking Performance through Multi-Task Learning Transformer
• Group-level Brain Decoding with Deep Learning
• Awake at the Wheel: Enhancing Automotive Safety through EEG-Based Fatigue Detection
• Collaborative Brain-Computer Interface for Human Interest Detection in Complex and Dynamic Settings
• I see artifacts: ICA-based EEG artifact removal does not improve deep network decoding across three BCI tasks
• Adversarial Artifact Detection in EEG-Based Brain-Computer Interfaces
• Q-EEGNet: an Energy-Efficient 8-bit Quantized Parallel EEGNet Implementation for Edge Motor-Imagery Brain--Machine Interfaces
• Towards Stroke Patients' Upper-limb Automatic Motor Assessment Using Smartwatches
• Feature Reweighting for EEG-based Motor Imagery Classification
• Analyzing P300 Distractors for Target Reconstruction
• KAM -- a Kernel Attention Module for Emotion Classification with EEG Data
• SSTAF: Spatial-Spectral-Temporal Attention Fusion Transformer for Motor Imagery Classification
• A Temporal-Spectral Fusion Transformer with Subject-Specific Adapter for Enhancing RSVP-BCI Decoding
• xEEGNet: Towards Explainable AI in EEG Dementia Classification
• Decoding Neural Activity to Assess Individual Latent State in Ecologically Valid Contexts
• Systematic Review of Experimental Paradigms and Deep Neural Networks for Electroencephalography-Based Cognitive Workload Detection
• D-FaST: Cognitive Signal Decoding with Disentangled Frequency-Spatial-Temporal Attention
• Neural Networks Meet Neural Activity: Utilizing EEG for Mental Workload Estimation
• Brain Topography Adaptive Network for Satisfaction Modeling in Interactive Information Access System
• MSA-CNN: A Lightweight Multi-Scale CNN with Attention for Sleep Stage Classification
• Exploring EEG and Eye Movement Fusion for Multi-Class Target RSVP-BCI
• vEEGNet: learning latent representations to reconstruct EEG raw data via variational autoencoders
• Cueless EEG imagined speech for subject identification: dataset and benchmarks
• Bayesian Time-Series Classifier for Decoding Simple Visual Stimuli from Intracranial Neural Activity
• Inter Subject Emotion Recognition Using Spatio-Temporal Features From EEG Signal
• A Fuzzy-based Approach to Predict Human Interaction by Functional Near-Infrared Spectroscopy
• ESI-GAL: EEG Source Imaging-based Trajectory Estimation for Grasp and Lift Task
• Comparative Analysis of Deep Learning Approaches for Harmful Brain Activity Detection Using EEG
• Front-end Replication Dynamic Window (FRDW) for Online Motor Imagery Classification
• Error-related Potential Variability: Exploring the Effects on Classification and Transferability
• Motor Imagery EEG Signal Classification Using Minimally Random Convolutional Kernel Transform and Hybrid Deep Learning
• Towards Interactive Reinforcement Learning with Intrinsic Feedback
• Classification of BCI-EEG based on augmented covariance matrix
• Enhancing Representation Learning of EEG Data with Masked Autoencoders
• Exploring Deep Learning Models for EEG Neural Decoding
• Adaptive Modality Balanced Online Knowledge Distillation for Brain-Eye-Computer based Dim Object Detection
• A dual-system approach for epilepsy diagnosis: integrating mamba-Bi-LSTM architecture with SHAP-based verification
• Spatial Imputation Drives Cross-Domain Alignment for EEG Classification
• NeuroSleepNet: An Explainable Multi-Head Attention-Based Framework With Spatial and Multi-Scale Independent Temporal Context Learning for Automatic Sleep Stage Scoring
• Cross-Subject Deep Transfer Models for Evoked Potentials in Brain-Computer Interface
• BRAVE: Brain-Controlled Prosthetic Arm with Voice Integration and Embodied Learning for Enhanced Mobility
• MIRepNet: A Pipeline and Foundation Model for EEG-Based Motor Imagery Classification
• A Combined Channel Approach for Decoding Intracranial EEG Signals: Enhancing Accuracy through Spatial Information Integration
• RISE-iEEG: Robust to Inter-Subject Electrodes Implantation Variability iEEG Classifier
• DiSCo: Diffusion Sequence Copilots for Shared Autonomy
• Closed loop BCI System for Cybathlon 2020
• SCVCNet: Sliding cross-vector convolution network for cross-task and inter-individual-set EEG-based cognitive workload recognition
• iFuzz-Meta: An Interpretable Fuzzy Learning Framework Bridging Top-Down and Bottom-Up Knowledge Integration
• Cognitive Load Estimation Using Brain Foundation Models and Interpretability for BCIs
• Masked Generative-Contrastive Representation Learning for Cross-Dataset EEG-Based Emotion Recognition
• Spatial Distillation based Distribution Alignment (SDDA) for Cross-Headset EEG Classification
• A Language-Guided Multimodal Foundation Model for Zero-Shot and Multi-Task Brain Signal Analysis
• CwA-T: A Channelwise AutoEncoder with Transformer for EEG Abnormality Detection