Subject-Invariant Deep Neural Networks Based on Baseline Correction for EEG Motor Imagery BCI

Electroencephalography (EEG)-based brain–computer interface (BCI) systems have been extensively used in various applications, such as communication, control, and rehabilitation. However, individual anatomical and physiological differences cause subject-specific variability of EEG signals for the same task, and BCI systems thus require a calibration procedure that adjusts system parameters to each subject. To overcome this problem, we propose a subject-invariant deep neural network (DNN) using baseline-EEG signals that can be recorded from subjects resting in comfortable states. We first modeled the deep features of EEG signals as a decomposition of subject-invariant and subject-variant features corrupted by anatomical/physiological characteristics. Subject-variant features were then removed from the deep features by learning the network with a baseline correction module (BCM) using the underlying individual information in baseline-EEG signals. The subject-invariant loss forces the BCM to assemble subject-invariant features that have the same class, irrespective of the subject. Using 1-min baseline-EEG signals of the new subject, our algorithm can eliminate subject-variant components from test data without the calibration process. The experimental results show that our subject-invariant DNN framework significantly increases decoding accuracies of the conventional DNN methods for BCI systems. Furthermore, feature visualizations illustrate that the proposed BCM extracts subject-invariant features that are close to each other in the same class.

[1]  Woo‐Jin Song,et al.  FGANet: fNIRS-Guided Attention Network for Hybrid EEG-fNIRS Brain-Computer Interfaces , 2022, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[2]  Xingyu Wang,et al.  Feature Extraction Method Based on Filter Banks and Riemannian Tangent Space in Motor-Imagery BCI , 2022, IEEE Journal of Biomedical and Health Informatics.

[3]  Hamid Fekri Azgomi,et al.  A Mixed Filtering Approach for Real-Time Seizure State Tracking Using Multi-Channel Electroencephalography Data , 2021, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[4]  F. Morabito,et al.  A Hybrid-Domain Deep Learning-Based BCI For Discriminating Hand Motion Planning From EEG Sources , 2021, Int. J. Neural Syst..

[5]  Byoung-Kyong Min,et al.  Deep-Learning-Based Automatic Selection of Fewest Channels for Brain–Machine Interfaces , 2021, IEEE Transactions on Cybernetics.

[6]  Cuntai Guan,et al.  MIN2Net: End-to-End Multi-Task Learning for Subject-Independent Motor Imagery EEG Classification , 2021, IEEE Transactions on Biomedical Engineering.

[7]  Xiaohui Liu,et al.  A review on transfer learning in EEG signal analysis , 2021, Neurocomputing.

[8]  Chun-Shu Wei,et al.  Low-Dimensional Subject Representation-Based Transfer Learning in EEG Decoding , 2020, IEEE Journal of Biomedical and Health Informatics.

[9]  Deniz Erdoğmuş,et al.  Disentangled Adversarial Autoencoder for Subject-Invariant Physiological Feature Extraction , 2020, IEEE Signal Processing Letters.

[10]  Klaus-Robert Müller,et al.  Enhanced Performance of a Brain Switch by Simultaneous Use of EEG and NIRS Data for Asynchronous Brain-Computer Interface , 2020, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[11]  Huiguang He,et al.  Multisource Transfer Learning for Cross-Subject EEG Emotion Recognition , 2020, IEEE Transactions on Cybernetics.

[12]  Silvestro Micera,et al.  Deep learning-based BCI for gait decoding from EEG with LSTM recurrent neural network , 2020, Journal of neural engineering.

[13]  Bangyan Zhou,et al.  To Explore the Potentials of Independent Component Analysis in Brain-Computer Interface of Motor Imagery , 2020, IEEE Journal of Biomedical and Health Informatics.

[14]  Ye Wang,et al.  Learning Invariant Representations From EEG via Adversarial Inference , 2020, IEEE Access.

[15]  Nadia Mammone,et al.  A deep CNN approach to decode motor preparation of upper limbs from time-frequency maps of EEG signals at source level , 2020, Neural Networks.

[16]  Natarajan Sriraam,et al.  EEG based multi-class seizure type classification using convolutional neural network and transfer learning , 2020, Neural Networks.

[17]  Lina Yao,et al.  Motor Imagery Classification via Temporal Attention Cues of Graph Embedded EEG Signals , 2020, IEEE Journal of Biomedical and Health Informatics.

[18]  Woo-Jin Song,et al.  Multilevel Feature Fusion With 3D Convolutional Neural Network for EEG-Based Workload Estimation , 2020, IEEE Access.

[19]  Gaël Varoquaux,et al.  Predictive regression modeling with MEG/EEG: from source power to signals and cognitive states , 2019, NeuroImage.

[20]  Seong-Whan Lee,et al.  Subject-Independent Brain–Computer Interfaces Based on Deep Convolutional Neural Networks , 2019, IEEE Transactions on Neural Networks and Learning Systems.

[21]  Heung-Il Suk,et al.  Mutual Information-Driven Subject-Invariant and Class-Relevant Deep Representation Learning in BCI , 2019, IEEE Transactions on Neural Networks and Learning Systems.

[22]  Kyung Joon Kwon,et al.  Multitask bilateral learning for real‐time image enhancement , 2019, Journal of the Society for Information Display.

[23]  Lina Yao,et al.  A Convolutional Recurrent Attention Model for Subject-Independent EEG Signal Analysis , 2019, IEEE Signal Processing Letters.

[24]  Gernot R. Müller-Putz,et al.  Domain Adaptation Techniques for EEG-Based Emotion Recognition: A Comparative Study on Two Public Datasets , 2019, IEEE Transactions on Cognitive and Developmental Systems.

[25]  Seong-Whan Lee,et al.  EEG dataset and OpenBMI toolbox for three BCI paradigms: an investigation into BCI illiteracy , 2019, GigaScience.

[26]  B. B. Zaidan,et al.  A review of disability EEG based wheelchair control system: Coherent taxonomy, open challenges and recommendations , 2018, Comput. Methods Programs Biomed..

[27]  P. Purdon,et al.  Electroencephalographic markers of brain development during sevoflurane anaesthesia in children up to 3 years old , 2018, British journal of anaesthesia.

[28]  Shuicheng Yan,et al.  Learning Temporal Information for Brain-Computer Interface Using Convolutional Neural Networks , 2018, IEEE Transactions on Neural Networks and Learning Systems.

[29]  S. Zafeiriou,et al.  ArcFace: Additive Angular Margin Loss for Deep Face Recognition , 2018, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[30]  Lina Yao,et al.  Multi-Person Brain Activity Recognition via Comprehensive EEG Signal Analysis , 2017, MobiQuitous.

[31]  Begoña García-Zapirain,et al.  Comparison of background EEG activity of different groups of patients with idiopathic epilepsy using Shannon spectral entropy and cluster-based permutation statistical testing , 2017, PloS one.

[32]  Phillip M. Alday.  How much baseline correction do we need in ERP research? Extended GLM model can replace baseline correction while lifting its limits. , 2017, Psychophysiology.

[33]  Qiang Gao,et al.  Noninvasive Electroencephalogram Based Control of a Robotic Arm for Writing Task Using Hybrid BCI System , 2017, BioMed research international.

[34]  Ugur Halici,et al.  A novel deep learning approach for classification of EEG motor imagery signals , 2017, Journal of neural engineering.

[35]  Brent Lance,et al.  EEGNet: a compact convolutional neural network for EEG-based brain–computer interfaces , 2016, Journal of neural engineering.

[36]  Kilian Q. Weinberger,et al.  Deep Feature Interpolation for Image Content Changes , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[37]  Yoshua Bengio,et al.  Online and offline handwritten Chinese character recognition: A comprehensive study and new benchmark , 2016, Pattern Recognit..

[38]  Bernhard Schölkopf,et al.  Transfer Learning in Brain-Computer Interfaces , 2015, IEEE Computational Intelligence Magazine.

[39]  Mohammed Yeasin,et al.  Learning Representations from EEG with Deep Recurrent-Convolutional Neural Networks , 2015, ICLR.

[40]  James Philbin,et al.  FaceNet: A unified embedding for face recognition and clustering , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[41]  Gian Domenico Iannetti,et al.  Single-trial time–frequency analysis of electrocortical signals: Baseline correction and beyond , 2014, NeuroImage.

[42]  Cuntai Guan,et al.  Filter Bank Common Spatial Pattern (FBCSP) in Brain-Computer Interface , 2008, 2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence).

[43]  M. Congedo,et al.  A review of classification algorithms for EEG-based brain–computer interfaces , 2007, Journal of neural engineering.

[44]  Masa-aki Sato,et al.  Hierarchical Bayesian estimation for MEG inverse problem , 2004, NeuroImage.

[45]  Gert Pfurtscheller,et al.  Motor imagery and direct brain-computer communication , 2001, Proc. IEEE.

[46]  J. Fermaglich.  Electric Fields of the Brain: The Neurophysics of EEG , 1982 .

[47]  Jost Tobias Springenberg,et al.  Supplementary for: Deep learning with convolutional neural networks for EEG decoding and visualization , 2017 .

[48]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[49]  Tara N. Sainath,et al.  Deep Neural Networks for Acoustic Modeling in Speech Recognition: The Shared Views of Four Research Groups , 2012, IEEE Signal Processing Magazine.

[50]  K.-R. Muller,et al.  Optimizing Spatial filters for Robust EEG Single-Trial Analysis , 2008, IEEE Signal Processing Magazine.