Algorithms benchmarking for removing EOG artifacts in brain computer interface
- 84 Downloads
In this paper, the author investigates and compares two types of algorithms for removing EOG artifacts in EEG measurements for brain computer interface applications. First, the author describes the sources that introduce EOG artifacts and briefly reviews the existing algorithms for EOG artifacts removal in EEG signals. Two selected algorithms—recursive least square (RLS) and blind source separation (BSS) are discussed in details. The algorithms were chosen either for its good performance or for its low computational charge to be used in ambulatory monitoring devices that are developed within our research program. In the third part, the experimental protocol to collect EOG and EEG measurements are described. The forth part summarizes the results after applying two algorithms to the collected EEG signals. From these results, we draw conclusions on the performance of these selected algorithms. Several possible improvements or challenges are proposed in the discussion.
KeywordsEOG artefacts EEG analysis Digital signal processing Recursive least square (RLS) Blind source separation (BSS) Brain computer interface
The authors would like to thank the Program for Professor of Special Appointment (Eastern Scholar) at Shanghai Institutions of Higher Learning (No. TP2015029) for financial support.
- 1.Atwood, H.L., MacKay, W.A.: Essentials of Neurophysiology, pp. 200–205. B.C. Decker, Hamilton (1993)Google Scholar
- 2.Tyner, F.S., Knott, J.R.: Fundamentals of EEG Technology, Volume 1: Basic Concepts and Methods. Raven press, New York (1989)Google Scholar
- 3.Niedermeyer, E., Lopes da Silva, F.H.: Electroencephalography: Basic Principles, Clinical Applications and Related Fields, 3rd edn, pp. 351–354. Lippincott, Williams & Wilkins, Philadelphia (1993)Google Scholar
- 11.Woestengurg, J.C., Verbaten, M.N., Slangen, J.L.: The removal of the eye movement artifact from the EEG by regression analysis in the frequency domain. Biol. Physiol. 16, 127–147 (1983)Google Scholar
- 19.Zhou, W., Gotman, J.: Removal of EMG and ECG artifacts from EEG based on wavelet transform and ICA. In: Proceedings of 26th Annual International Conference IEEE Engineering in Medicine and Biology Society (EMBS04), San Francisco, CA, Sep. 15, pp. 392–395 (2004)Google Scholar
- 20.Nguyen, H.-A.T., Musson, J., Li, J., McKenzie, F., Zhang, G., Zu, R., Richey, C., Schell, T. EEG artifact removal using a Wavelet Neural Network. In Proceedings of Mod Sim World (2010)Google Scholar
- 21.Jung, T.-P., Humphries, C., Lee, T.-W., Makeig, S., McKeown, M.J., Iragui, V., Sejnowski, T.J.: Extended ICA removes artifacts from electroencephalographic recordings. Adv. Neural Inf. Process. Syst. 10, 894–900 (1998)Google Scholar
- 27.Rowland, V.: Cortical steady potential (direct current potential) in reinforcement and learning. Progr. Physiol. Psychol. 2, 1–77 (1968)Google Scholar