Abstract
Electroencephalography (EEG) is a non-invasive technique to record brain activities in natural settings. Ocular Artifacts (OA) usually contaminates EEG signals, removal of which is critical for accurate feature extraction and classification. With the increasing adoption of wearable technologies, single-channel real-time EEG systems that often require real-time signal processing for immediate real-time feedback are becoming more prevalent. However, traditional OA removal algorithms usually require multiple channels of EEG data, are computationally expensive, and do not perform well in real-time. In this article, a new hybrid algorithm is proposed that autonomously detects OA and subsequently removes OA from a single-channel steaming EEG data in real-time. The proposed single EEG channel algorithm also does not require additional reference electrooculography (EOG) channel. The algorithm has also been implemented on an embedded hardware platform of single channel wearable EEG system (NeuroMonitor). The algorithm first detects the OA zones using an Algebraic approach and then removes these artifacts from the detected OA zones using the Discrete Wavelet Transform (DWT) decomposition method. The de-noising technique is applied only to the OA zone, which minimizes loss of neural information outside the OA zone. A qualitative and quantitative performance evaluation was carried out with a 0.5s epoch in overlapping sliding window technique using time-frequency analysis, mean square coherence, and correlation coefficient statistics. The hybrid OA removal algorithm demonstrated real-time operation with 3s latency on the PSoC-3-microcontroller-based EEG system. Successful implementation of OA removal from single-channel real-time EEG data using the proposed algorithm shows promise for real-time feedback applications of wearable EEG devices.
- Raymundo Cassani, Tiago H. Falk, Francisco J. Fraga, Paulo A. M. Kanda, and Renato Anghinah. 2014. The effects of automated artifact removal algorithms on electroencephalography-based Alzheimer's disease diagnosis. Front Aging Neurosci. 6, Article 55, (March 2014), 13.Google Scholar
- Yu M. Chi, Patrick Ng, and Gert Cauwenberghs. 2013. Wireless noncontact ECG and EEG biopotential sensors. ACM Trans. Embedd. Comput. Syst. 12, 4, Article 103 (June 2013), 19. Google Scholar
Digital Library
- Sergi Consul-Pacareu, R. Mahajan, N. Sahadat, and B. I. Morshed. 2014. Wearable ambulatory 2-channel EEG NeuroMonitor platform for real-life engagement monitoring based on brain activities at the prefrontal cortex. In Proceedings of the 2014 IAJC-ISAM International Conference, Paper 78, 12.Google Scholar
- A. Delorme and S. Makeig. 2004. EEGLAB: An open source toolbox for analysis of single-trial EEG dynamics. J Neurosci. Meth. 134 (2004), 9--21.Google Scholar
- H. T. Gorji, A. Koohpayezadeh, and J. Haddadnia. 2013. Ocular artifact detection and removing from EEG by wavelet families: A comparative study. J. Inf. Engin. Appl. 3 (2013), 39--47.Google Scholar
- C. A. Joyce, I. F. Gorodnitsky, and M. Kutas. 2004. Automatic removal of eye movement and blink artifacts from EEG data using blind component separation. Psychophysiology 41 (2004), 313--325.Google Scholar
Cross Ref
- Saleha Khatun, Ruhi Mahajan, and Bashir I. Morshed. 2015. Comparative analysis of wavelet-based approaches for reliable removal of ocular artifacts from single channel EEG. In Proceedings of the IEEE Electro/Information Technology (EIT), 335--340.Google Scholar
- V. Krishnaveni, S. Jayaraman, L. Anitha, and K. Ramadoss. 2006. Removal of ocular artifacts from EEG using adaptive thresholding of wavelet coefficients. J. Neur. Engin. 3 (2006), 338.Google Scholar
Cross Ref
- P. S. Kumar, R. Arumuganathan, K. Sivakumar, and C. Vimal. 2008. A wavelet-based statistical method for de-noising of ocular artifacts in EEG signals. Int. J. Comput. Sci. Netw. Secur. 8 (2008), 87--92.Google Scholar
- R. Lloyd, R. Goulding, P. Filan, and G. Boylan. 2015. Overcoming the practical challenges of electroencephalography for very preterm infants in the neonatal intensive care unit. Acta Paed. 104 (2015), 152--157.Google Scholar
Cross Ref
- Ruhi Mahajan, C. A. Majmudar, S. Khatun, B. I. Morshed, and G. M. Bidelman. 2014. NeuroMonitor ambulatory EEG device: Comparative analysis and its application for cognitive load assessment. In Proceedings of IEEE Healthcare Innovations and Point-of-Care Technologies Conference, 133--136.Google Scholar
- Ruhi Mahajan and B. I. Morshed. 2015. Unsupervised eye blink artifact denoising in EEG data with modified multiscale sample entropy, kurtosis and wavelet-ICA. IEEE J. Biomed. Health Inform. 19, 1 (Jan. 2015), 158--165.Google Scholar
Cross Ref
- Charvi A. Majmudar, Ruhi Mahajan, and Bashir I. Morshed. 2015. Real-time hybrid ocular artifact detection and removal for single channel EEG. In Proceedings of IEEE Intl Conf Electro/Information Technology (EIT), 330--334.Google Scholar
- Fabien Masse, Martien van Vussel, Aline Serteyn, Johan Arends, and Julien Penders. 2013. Miniaturized wireless ECG monitor for real-time detection of epileptic seizures. ACM Trans. Embedd. Comput. Syst. 12, 4, Article 102 (June 2013), 21. Google Scholar
Digital Library
- M. Mboup. 2012. Neuronal spike detection and localisation via Volterra filtering. In Proceedings of the IEEE International. Workshop on Machine Learning for Signal Processing. 1--5.Google Scholar
- B. I. Morshed and A. Khan. 2014. A brief review of technologies and challenges to monitor brain activities. J. Bioengin. Biomed. Sci. 4, 1 (2014), 1--10.Google Scholar
Cross Ref
- Yunyong Nam, Seungmin Rho, and Chulung Lee. 2013. Physical activity recognition using multiple sensors embedded in a wearable device. ACM Trans. Embedded. Comput. Syst. 12, 2, Article 26 (Feb. 2013), 14. Google Scholar
Digital Library
- Borna Noureddin. 2010. Online Removal of Eye Movement and Blink Artifacts from EEG Signals without EOG. Ph.D. Dissertation. The University of British Columbia, Vancouver, B.C., Canada.Google Scholar
- Hong Peng, Dennis Majoe, and Thomas Kaegi-Trachsel. 2011. Design and application of a novel wearable EEG system for e-healthcare. In Proceedings of the International Workshop on Ubiquitous Affective Awareness and Intelligent Interaction. ACM, New York, NY, 1--8. Google Scholar
Digital Library
- S. V. Ramanan, N. Kalpakam, and J. Sahambi. 2004. A novel wavelet-based technique for detection and de-noising of ocular artifact in normal and epileptic electroencephalogram. In Proceedings of the Brain Inspired Cognitive Systems Conference. 2, 1027--1031.Google Scholar
- M. H. Soomro, N. Badruddin, M. Z. Yusoff, and M. A. Jatoi. 2013. Automatic eye-blink artifact removal method based on EMD-CCA. In Proceedings of the ICME International Conference on Complex Medical Engineering (CME). 186--190.Google Scholar
- K. T. Sweeney, S. F. McLoone, and T. E. Ward. 2013. The use of ensemble empirical mode decomposition with canonical correlation analysis as a novel artifact removal technique. IEEE Trans. Biomed. Engin. 60, (2013), 97--105.Google Scholar
- M. Teplan. 2002. Fundamentals of EEG measurement. Measurement Sci. Rev., 2, (2002), 1--11.Google Scholar
- Z. Tiganj, M. Mboup, C. Pouzat, and L. Belkoura. 2010. An algebraic method for eye blink artifacts detection in single channel EEG recordings. In Proceedings of the 17th International Conference on Biomagnetism Advances in Biomagnetism--Biomag. 175--178.Google Scholar
Index Terms
Autonomous OA Removal in Real-Time from Single Channel EEG Data on a Wearable Device Using a Hybrid Algebraic-Wavelet Algorithm
Recommendations
Employing spatially constrained ICA and wavelet denoising, for automatic removal of artifacts from multichannel EEG data
Detecting artifacts produced in electroencephalographic (EEG) data by muscle activity, eye blinks and electrical noise, etc., is an important problem in EEG signal processing research. These artifacts must be corrected before further analysis because it ...
EOG artifact removal using a wavelet neural network
In this paper, we developed a wavelet neural network (WNN) algorithm for electroencephalogram (EEG) artifact. The algorithm combines the universal approximation characteristics of neural networks and the time/frequency property of wavelet transform, ...
An epileptic seizure prediction algorithm from scalp EEG based on morphological filter and kolmogorov complexity
ICDHM'07: Proceedings of the 1st international conference on Digital human modelingEpilepsy is the most common neurological disorder in the world, second only to stroke. There are nearly 15 million patients suffer from refractory epilepsy, with no available therapy. Although most seizures are not life threatening, they are an ...






Comments