A Novel Steady-State Visually Evoked Potential (SSVEP) Based Brain Computer Interface Paradigm for Disabled Individuals

This study provides an insight into a novel steady state visually evoked potential (SSVEP) brain computer interface (BCI) approach. In this approach, four groups of light emitting diodes (LEDs) that flicker at different frequencies are used and each of these groups consist of three LEDs connected in series. By providing visual attention to these LEDs, corresponding electroencephalograph (EEG) signals were obtained in the visual cortex area of the brain. Using suitable signal processing algorithms, acquired EEG signals were classified at different frequencies and given as inputs to a brain computer interface system that can control the movement of a wheelchair. This method provides a platform for individuals who are affected by neuromuscular degenerative diseases (NMD) such as Amyotrophic Lateral Sclerosis (ALS), Locked-in Syndrome (LIS) etc, to help them lead an independent life. Two different SSVEP approaches were carried out on four healthy subjects for prototype testing. First approach was based on four groups of LEDs flickering at different frequencies ranging from 7 Hz to 15 Hz and the subjects selectively paid attention to one group of LEDs at a time. The second approach was based on simultaneous flickering of two groups of LEDs at different frequency combinations. Five trials were conducted on four subjects to test the performance of the system. The average accuracy obtained with each of the methods was greater than 70% with an average time of less than 10 seconds to trigger a command for BCI based application. The proposed system can thus provide a visual stimulator based on simple and customizable LED for a cost- effective BCI approach. Also, the efficiency and accuracy of the proposed SSVEP approach was compared to audio steady state response (ASSR) approach, where the subjects concentrated to two tones of beat frequencies at 37 Hz and 43 Hz. The average accuracy obtained with ASSR approach was only 47.5% with an average time of 18.72 seconds. Compared to ASSR, SSVEP approach is 23% more efficient.

[1]  K. M. Donnelly,et al.  Quantitative analysis of the EEG posterior-dominant rhythm in healthy adolescents , 2008, Clinical Neurophysiology.

[2]  Satoru Goto,et al.  Characteristic analysis of visual evoked potentials and posterior dominant rhythm by use of EEG model , 2013, 2013 13th International Conference on Control, Automation and Systems (ICCAS 2013).

[3]  Ivan Volosyak,et al.  Toward BCI Wizard - best BCI approach for each user , 2010, 2010 Annual International Conference of the IEEE Engineering in Medicine and Biology.

[4]  Wei Wu,et al.  Frequency Recognition Based on Canonical Correlation Analysis for SSVEP-Based BCIs , 2006, IEEE Transactions on Biomedical Engineering.

[5]  Brendan Z. Allison,et al.  How Many People Could Use an SSVEP BCI? , 2012, Front. Neurosci..

[6]  Héctor Pomares,et al.  Evidences of cognitive effects over auditory steady-state responses by means of artificial neural networks and its use in brain-computer interfaces , 2009, Neurocomputing.

[7]  H. Levent Akin,et al.  An SSVEP based BCI to control a humanoid robot by using portable EEG device , 2013, 2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

[8]  Bin He,et al.  BRAIN^COMPUTER INTERFACE , 2007 .

[9]  Sandeep Mittal,et al.  Invasive electroencephalography monitoring: Indications and presurgical planning , 2014, Annals of Indian Academy of Neurology.

[10]  Chang-Hwan Im,et al.  A vision-free brain-computer interface (BCI) paradigm based on auditory selective attention , 2011, 2011 Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[11]  H. Jasper,et al.  The ten-twenty electrode system of the International Federation. The International Federation of Clinical Neurophysiology. , 1999, Electroencephalography and clinical neurophysiology. Supplement.