TY - GEN
T1 - Brain Computer Interface Switch Based on Quasi-Steady-State Visual Evoked Potentials
AU - Kaya, Ibrahim
AU - Bohorquez, Jorge E.
AU - Ozdamar, Ozcan
N1 - Publisher Copyright:
© 2019 IEEE.
PY - 2019/5/16
Y1 - 2019/5/16
N2 - Gaze detection by Steady State Visually Evoked Potentials (SSVEPs) has been a very popular topic for neural rehabilitation and especially Brain Computer Interface (BCI) research. Visual Evoked Potentials (VEPs) provide reliable robust electrophysiological signals for communication and control applications. In this research Quasi-Steady-State VEPs (QSS-VEPs) and their correspondence in this kind of dual target gaze detection application has been demonstrated. Methods for a dual target BCI switch has been developed. Since QSS-VEPs enable acquisition of both SSVEP and transient VEP (TR-VEP) signals at the same time, exploitation of them in a two target BCI application was reasonable. Stimulations of low and high rate pattern-reversal VEPs were utilized in gaze detection. QSS-VEPs and transient VEPs were compared. Receiver Operating Characteristics (ROC) curves were calculated in order to assess performance. We have found that it is possible to achieve high accuracy or ROC area values with QSS-VEPs. The best performance was obtained at 50 reversal per second (rps) stimulation rate. At this rate deconvolved transient signals, which normally has low performance compared to QSS-VEPs, resulted in comparable Information Transfer Rates (ITR) with QSS-VEPs.
AB - Gaze detection by Steady State Visually Evoked Potentials (SSVEPs) has been a very popular topic for neural rehabilitation and especially Brain Computer Interface (BCI) research. Visual Evoked Potentials (VEPs) provide reliable robust electrophysiological signals for communication and control applications. In this research Quasi-Steady-State VEPs (QSS-VEPs) and their correspondence in this kind of dual target gaze detection application has been demonstrated. Methods for a dual target BCI switch has been developed. Since QSS-VEPs enable acquisition of both SSVEP and transient VEP (TR-VEP) signals at the same time, exploitation of them in a two target BCI application was reasonable. Stimulations of low and high rate pattern-reversal VEPs were utilized in gaze detection. QSS-VEPs and transient VEPs were compared. Receiver Operating Characteristics (ROC) curves were calculated in order to assess performance. We have found that it is possible to achieve high accuracy or ROC area values with QSS-VEPs. The best performance was obtained at 50 reversal per second (rps) stimulation rate. At this rate deconvolved transient signals, which normally has low performance compared to QSS-VEPs, resulted in comparable Information Transfer Rates (ITR) with QSS-VEPs.
KW - BCI Switch
KW - Brain-computer Interfaces (BCI)
KW - Quasi-Steady-State VEP (QSS-VEP)
KW - Steady-State VEP (SSVEP)
KW - Transient VEP
KW - Visual Evoked Potential
UR - http://www.scopus.com/inward/record.url?scp=85066767091&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85066767091&partnerID=8YFLogxK
U2 - 10.1109/NER.2019.8716894
DO - 10.1109/NER.2019.8716894
M3 - Conference contribution
AN - SCOPUS:85066767091
T3 - International IEEE/EMBS Conference on Neural Engineering, NER
SP - 1175
EP - 1178
BT - 9th International IEEE EMBS Conference on Neural Engineering, NER 2019
PB - IEEE Computer Society
T2 - 9th International IEEE EMBS Conference on Neural Engineering, NER 2019
Y2 - 20 March 2019 through 23 March 2019
ER -