Automatic ocular artifacts removal in EEG using deep learning

Abstract Ocular artifacts (OAs) are one the most important form of interferences in the analysis of electroencephalogram (EEG) research. OAs removal/reduction is a key analysis before the processing of EEG signals. For classic OAs removal methods, either an additional electrooculogram (EOG) recording or multi-channel EEG is required. To address these limitations of existing methods, this paper investigates the use of deep learning network (DLN) to remove OAs in EEG signals. The proposed method consists of offline stage and online stage. In the offline stage, training samples without OAs are intercepted and used to train an DLN to reconstruct the EEG signals. The high-order statistical moments information of EEG is therefore learned. In the online stage, the trained DLN is used as a filter to automatically remove OAs from the contaminated EEG signals. Compared with the exiting methods, the proposed method has the following advantages: (i) nonuse of additional EOG reference signals, (ii) any few number of EEG channels can be analyzed, (iii) time saving, and (iv) the strong generalization ability, etc. In this paper, both public database and lab individual data for EEG analysis are used, we compared the proposed method with the classic independent component analysis (ICA), kurtosis-ICA (K-ICA), Second-order blind identification (SOBI) and a shallow network method. Experimental results show that the proposed method performs better even for very noisy EEG.

[1]  L. Sherlin,et al.  Chapter 4 – Diagnosing and treating brain function through the use of low resolution brain electromagnetic tomography (LORETA) , 2009 .

[2]  Bozenna Pasik-Duncan,et al.  EEG ocular artifact removal through ARMAX model system identification using extended least squares , 2003, Commun. Inf. Syst..

[3]  Atilla Kilicarslan,et al.  A robust adaptive denoising framework for real-time artifact removal in scalp EEG measurements , 2016, Journal of neural engineering.

[4]  V. Krishnaveni,et al.  Removal of ocular artifacts from EEG using adaptive thresholding of wavelet coefficients , 2006, Journal of neural engineering.

[5]  Jiang Li,et al.  EOG artifact removal using a wavelet neural network , 2012, Neurocomputing.

[6]  Gleb V. Tcheslavski,et al.  Independent Component Analysis for EOG artifacts minimization of EEG signals using kurtosis as a threshold , 2015, 2015 2nd International Conference on Electrical Information and Communication Technologies (EICT).

[7]  Müjdat Çetin,et al.  A latent discriminative model-based approach for classification of imaginary motor tasks from EEG data , 2012, Journal of neural engineering.

[8]  G. Wilson,et al.  Removal of ocular artifacts from electro-encephalogram by adaptive filtering , 2004, Medical and Biological Engineering and Computing.

[9]  Panagiotis D. Bamidis,et al.  REG-ICA: A hybrid methodology combining Blind Source Separation and regression techniques for the rejection of ocular artifacts , 2011, Biomed. Signal Process. Control..

[10]  Trieu Phat Luu,et al.  Gait adaptation to visual kinematic perturbations using a real-time closed-loop brain–computer interface to a virtual reality avatar , 2016, Journal of neural engineering.

[11]  V. A. Makarov,et al.  Recovering EEG brain signals: Artifact suppression with wavelet enhanced independent component analysis , 2006, Journal of Neuroscience Methods.

[12]  M L Good,et al.  Educational simulation of the electroencephalogram (EEG). , 2001, Technology and health care : official journal of the European Society for Engineering and Medicine.

[13]  K. P. Indiradevi,et al.  A multi-level wavelet approach for automatic detection of epileptic spikes in the electroencephalogram , 2008, Comput. Biol. Medicine.

[14]  Daniel P Ferris,et al.  Isolating gait-related movement artifacts in electroencephalography during human walking , 2015, Journal of neural engineering.

[15]  F. La Foresta,et al.  Automatic Artifact Rejection From Multichannel Scalp EEG by Wavelet ICA , 2012, IEEE Sensors Journal.

[16]  Raveendran Paramesran,et al.  Artifacts-matched blind source separation and wavelet transform for multichannel EEG denoising , 2015, Biomed. Signal Process. Control..

[17]  Keum-Shik Hong,et al.  Linear parameter-varying model and adaptive filtering technique for detecting neuronal activities: an fNIRS study , 2013, Journal of neural engineering.

[18]  T. Sejnowski,et al.  Removing electroencephalographic artifacts by blind source separation. , 2000, Psychophysiology.

[19]  Fabio A. González,et al.  Supervised Greedy Layer-Wise Training for Deep Convolutional Networks with Small Datasets , 2015, ICCCI.

[20]  Klaus-Robert Müller,et al.  The non-invasive Berlin Brain–Computer Interface: Fast acquisition of effective performance in untrained subjects , 2007, NeuroImage.

[21]  Mahesh D. Pandey,et al.  Modified Cross-Correlation Method for the Blind Identification of Structures , 2010 .

[22]  Karim Faez,et al.  Comparison of Different Linear Filter Design Methods for Handling Ocular Artifacts in Brain Computer Interface System , 2014 .

[23]  C. Joyce,et al.  Automatic removal of eye movement and blink artifacts from EEG data using blind component separation. , 2004, Psychophysiology.

[24]  Arjon Turnip,et al.  Removal artifacts from EEG signal using independent component analysis and principal component analysis , 2014, 2014 2nd International Conference on Technology, Informatics, Management, Engineering & Environment.

[25]  Yufei Huang,et al.  A Deep Learning method for classification of images RSVP events with EEG data , 2013, 2013 IEEE Global Conference on Signal and Information Processing.

[26]  Banghua Yang,et al.  Fast removal of ocular artifacts from electroencephalogram signals using spatial constraint independent component analysis based recursive least squares in brain-computer interface , 2015, Frontiers of Information Technology & Electronic Engineering.

[27]  Justin A. Blanco,et al.  Modeling electroencephalography waveforms with semi-supervised deep belief nets: fast classification and anomaly measurement , 2011, Journal of neural engineering.

[28]  Luz María Alonso-Valerdi,et al.  Perception and Cognition of Cues Used in Synchronous Brain–Computer Interfaces Modify Electroencephalographic Patterns of Control Tasks , 2015, Front. Hum. Neurosci..

[29]  D. Pizzagalli Electroencephalography and High-Density Electrophysiological Source Localization , 2007 .

[30]  Eric Moulines,et al.  A blind source separation technique using second-order statistics , 1997, IEEE Trans. Signal Process..

[31]  Dirk Hagemann,et al.  The effects of ocular artifacts on (lateralized) broadband power in the EEG , 2001, Clinical Neurophysiology.

[32]  Jing Hu,et al.  Removal of EOG and EMG artifacts from EEG using combination of functional link neural network and adaptive neural fuzzy inference system , 2015, Neurocomputing.

[33]  Brian Litt,et al.  Semi-Supervised Anomaly Detection for EEG Waveforms Using Deep Belief Nets , 2010, 2010 Ninth International Conference on Machine Learning and Applications.

[34]  R. Ward,et al.  EMG and EOG artifacts in brain computer interface systems: A survey , 2007, Clinical Neurophysiology.

[35]  Thomas Hofmann,et al.  Greedy Layer-Wise Training of Deep Networks , 2007 .

[36]  Stephanie Brandl,et al.  Robust artifactual independent component classification for BCI practitioners , 2014, Journal of neural engineering.

[37]  Abbas Erfanian,et al.  A fully automatic ocular artifact suppression from EEG data using higher order statistics: improved performance by wavelet analysis. , 2010, Medical engineering & physics.

[38]  L. R. Quitadamo,et al.  Investigating the effects of a sensorimotor rhythm-based BCI training on the cortical activity elicited by mental imagery , 2014, Journal of neural engineering.

[39]  Maarten De Vos,et al.  P300 speller BCI with a mobile EEG system: comparison to a traditional amplifier , 2014, Journal of neural engineering.

[40]  Barak A. Pearlmutter,et al.  Blind source separation of multichannel neuromagnetic responses , 2000, Neurocomputing.

[41]  Yee Whye Teh,et al.  A Fast Learning Algorithm for Deep Belief Nets , 2006, Neural Computation.

[42]  Muhammad A. Kamran,et al.  Hybrid ICA—Regression: Automatic Identification and Removal of Ocular Artifacts from Electroencephalographic Signals , 2016, Front. Hum. Neurosci..

[43]  R. Sameni,et al.  An iterative subspace denoising algorithm for removing electroencephalogram ocular artifacts , 2014, Journal of Neuroscience Methods.

[44]  M. De Vos,et al.  Automated artifact removal as preprocessing refines neonatal seizure detection , 2011, Clinical Neurophysiology.

[45]  G. Birch,et al.  Time-frequency Analysis of Eye Blinks and Saccades in EOG for EEG Artifact Removal , 2007, 2007 3rd International IEEE/EMBS Conference on Neural Engineering.

[46]  Chao-Hung Lin,et al.  Wavelet-based envelope features with automatic EOG artifact removal: Application to single-trial EEG data , 2012, Expert Syst. Appl..

[47]  J P Rosenfeld,et al.  The Effects of Alpha (10-Hz) and Beta (22-Hz) “Entrainment” Stimulation on the Alpha and Beta EEG Bands: Individual Differences Are Critical to Prediction of Effects , 1997, Applied psychophysiology and biofeedback.

[48]  Joong Hoon Lee,et al.  CNT/PDMS-based canal-typed ear electrodes for inconspicuous EEG recording. , 2014, Journal of neural engineering.