Automatic ocular artifacts removal in EEG using deep learning

• Published in: Biomedical signal processing and control Vol. 43; pp. 148 - 158
• Main Authors: Yang, Banghua, Duan, Kaiwen, Fan, Chengcheng, Hu, Chenxiao, Wang, Jinlong
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.05.2018
• Subjects: Deep learning network (DLN); Electroencephalogram (EEG); Independent component analysis (ICA); Ocular artifacts (OAs) removal; Shallow network; Deep learning network (DLN); Electroencephalogram (EEG); Ocular artifacts (OAs) removal; Independent component analysis (ICA); Shallow network
• ISSN: 1746-8094; 1746-8108
• DOI: 10.1016/j.bspc.2018.02.021

•This paper investigates the use of deep learning network (DLN) to remove ocular artifacts in EEG signals.•For practical use of rehabilitation system based on brain computer interface, the method does not require additional EOG channels for recording additional EOG reference signals. Meanwhile, the ocular artifacts can be denoised without influence.•With few number of EEG analyzing channels and low time cost, combining without additional EOG channel recording, the method could be used in practical motor imagery rehabilitation system, which avoid the trivial preparation and experiment and more susceptible for patients.•In this paper, the proposed method is proved to have strong generalization ability after cross-subject testing. 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.