Publications
2015
Jin, Jing; Daly, Ian; Zhang, Yu; Wang, Xingyu; Cichocki, Andrzej
A new hybrid BCI paradigm based on P300 and SSVEP Journal Article
In: Journal of Neural Engineering, vol. 244, pp. 16–25, 2015.
Abstract | Links | BibTeX | Tags: BCI, Event-related potential, Hybrid BCI, P300, SSVEP
@article{Wang2014,
title = {A new hybrid BCI paradigm based on P300 and SSVEP},
author = {Jing Jin and Ian Daly and Yu Zhang and Xingyu Wang and Andrzej Cichocki},
url = {http://www.sciencedirect.com/science/article/pii/S016502701400209X},
doi = {doi:10.1016/j.jneumeth.2014.06.003},
year = {2015},
date = {2015-04-15},
journal = {Journal of Neural Engineering},
volume = {244},
pages = {16–25},
abstract = {Background
P300 and steady-state visual evoked potential (SSVEP) approaches have been widely used for brain–computer interface (BCI) systems. However, neither of these approaches can work for all subjects. Some groups have reported that a hybrid BCI that combines two or more approaches might provide BCI functionality to more users. Hybrid P300/SSVEP BCIs have only recently been developed and validated, and very few avenues to improve performance have been explored.
New method
The present study compares an established hybrid P300/SSVEP BCIs paradigm to a new paradigm in which shape changing, instead of color changing, is adopted for P300 evocation to decrease the degradation on SSVEP strength.
Result
The result shows that the new hybrid paradigm presented in this paper yields much better performance than the normal hybrid paradigm.
Comparison with existing method
A performance increase of nearly 20% in SSVEP classification is achieved using the new hybrid paradigm in comparison with the normal hybrid paradigm. All the paradigms except the normal hybrid paradigm used in this paper obtain 100% accuracy in P300 classification.
Conclusions
The new hybrid P300/SSVEP BCIs paradigm in which shape changing, instead of color changing, could obtain as high classification accuracy of SSVEP as the traditional SSVEP paradigm and could obtain as high classification accuracy of P300 as the traditional P300 paradigm. P300 did not interfere with the SSVEP response using the new hybrid paradigm presented in this paper, which was superior to the normal hybrid P300/SSVEP paradigm.},
keywords = {BCI, Event-related potential, Hybrid BCI, P300, SSVEP},
pubstate = {published},
tppubtype = {article}
}
Background
P300 and steady-state visual evoked potential (SSVEP) approaches have been widely used for brain–computer interface (BCI) systems. However, neither of these approaches can work for all subjects. Some groups have reported that a hybrid BCI that combines two or more approaches might provide BCI functionality to more users. Hybrid P300/SSVEP BCIs have only recently been developed and validated, and very few avenues to improve performance have been explored.
New method
The present study compares an established hybrid P300/SSVEP BCIs paradigm to a new paradigm in which shape changing, instead of color changing, is adopted for P300 evocation to decrease the degradation on SSVEP strength.
Result
The result shows that the new hybrid paradigm presented in this paper yields much better performance than the normal hybrid paradigm.
Comparison with existing method
A performance increase of nearly 20% in SSVEP classification is achieved using the new hybrid paradigm in comparison with the normal hybrid paradigm. All the paradigms except the normal hybrid paradigm used in this paper obtain 100% accuracy in P300 classification.
Conclusions
The new hybrid P300/SSVEP BCIs paradigm in which shape changing, instead of color changing, could obtain as high classification accuracy of SSVEP as the traditional SSVEP paradigm and could obtain as high classification accuracy of P300 as the traditional P300 paradigm. P300 did not interfere with the SSVEP response using the new hybrid paradigm presented in this paper, which was superior to the normal hybrid P300/SSVEP paradigm.
P300 and steady-state visual evoked potential (SSVEP) approaches have been widely used for brain–computer interface (BCI) systems. However, neither of these approaches can work for all subjects. Some groups have reported that a hybrid BCI that combines two or more approaches might provide BCI functionality to more users. Hybrid P300/SSVEP BCIs have only recently been developed and validated, and very few avenues to improve performance have been explored.
New method
The present study compares an established hybrid P300/SSVEP BCIs paradigm to a new paradigm in which shape changing, instead of color changing, is adopted for P300 evocation to decrease the degradation on SSVEP strength.
Result
The result shows that the new hybrid paradigm presented in this paper yields much better performance than the normal hybrid paradigm.
Comparison with existing method
A performance increase of nearly 20% in SSVEP classification is achieved using the new hybrid paradigm in comparison with the normal hybrid paradigm. All the paradigms except the normal hybrid paradigm used in this paper obtain 100% accuracy in P300 classification.
Conclusions
The new hybrid P300/SSVEP BCIs paradigm in which shape changing, instead of color changing, could obtain as high classification accuracy of SSVEP as the traditional SSVEP paradigm and could obtain as high classification accuracy of P300 as the traditional P300 paradigm. P300 did not interfere with the SSVEP response using the new hybrid paradigm presented in this paper, which was superior to the normal hybrid P300/SSVEP paradigm.
2013
Daly, Ian; Billinger, Martin; Laparra-Hernandez, Jose; Aloise, Fabio; Garcia, Mariano Lloria; Faller, Josef; Scherer, Reinhold; Muller-Putz, Gernot
On the control of Brain-computer interfaces by users with Cerebral palsy Journal Article
In: Clinical Neurophysiology, vol. 124, no. 9, pp. 1787-1797, 2013.
Abstract | Links | BibTeX | Tags: BCI, Cerebral palsy, ERD, Motor imagery, SSVEP
@article{Daly2013cpBCI,
title = {On the control of Brain-computer interfaces by users with Cerebral palsy},
author = {Ian Daly and Martin Billinger and Jose Laparra-Hernandez and Fabio Aloise and Mariano Lloria Garcia and Josef Faller and Reinhold Scherer and Gernot Muller-Putz},
url = {http://www.iandaly.co.uk/wp-content/uploads/2016/01/draft_6-0.pdf},
doi = {10.1016/j.clinph.2013.02.118},
year = {2013},
date = {2013-09-01},
journal = {Clinical Neurophysiology},
volume = {124},
number = {9},
pages = {1787-1797},
abstract = {OBJECTIVE:
Brain-computer interfaces (BCIs) have been proposed as a potential assistive device for individuals with cerebral palsy (CP) to assist with their communication needs. However, it is unclear how well-suited BCIs are to individuals with CP. Therefore, this study aims to investigate to what extent these users are able to gain control of BCIs.
METHODS:
This study is conducted with 14 individuals with CP attempting to control two standard online BCIs (1) based upon sensorimotor rhythm modulations, and (2) based upon steady state visual evoked potentials.
RESULTS:
Of the 14 users, 8 are able to use one or other of the BCIs, online, with a statistically significant level of accuracy, without prior training. Classification results are driven by neurophysiological activity and not seen to correlate with occurrences of artifacts. However, many of these users' accuracies, while statistically significant, would require either more training or more advanced methods before practical BCI control would be possible.
CONCLUSIONS:
The results indicate that BCIs may be controlled by individuals with CP but that many issues need to be overcome before practical application use may be achieved.
SIGNIFICANCE:
This is the first study to assess the ability of a large group of different individuals with CP to gain control of an online BCI system. The results indicate that six users could control a sensorimotor rhythm BCI and three a steady state visual evoked potential BCI at statistically significant levels of accuracy (SMR accuracies; mean ± STD, 0.821 ± 0.116, SSVEP accuracies; 0.422 ± 0.069).},
keywords = {BCI, Cerebral palsy, ERD, Motor imagery, SSVEP},
pubstate = {published},
tppubtype = {article}
}
OBJECTIVE:
Brain-computer interfaces (BCIs) have been proposed as a potential assistive device for individuals with cerebral palsy (CP) to assist with their communication needs. However, it is unclear how well-suited BCIs are to individuals with CP. Therefore, this study aims to investigate to what extent these users are able to gain control of BCIs.
METHODS:
This study is conducted with 14 individuals with CP attempting to control two standard online BCIs (1) based upon sensorimotor rhythm modulations, and (2) based upon steady state visual evoked potentials.
RESULTS:
Of the 14 users, 8 are able to use one or other of the BCIs, online, with a statistically significant level of accuracy, without prior training. Classification results are driven by neurophysiological activity and not seen to correlate with occurrences of artifacts. However, many of these users' accuracies, while statistically significant, would require either more training or more advanced methods before practical BCI control would be possible.
CONCLUSIONS:
The results indicate that BCIs may be controlled by individuals with CP but that many issues need to be overcome before practical application use may be achieved.
SIGNIFICANCE:
This is the first study to assess the ability of a large group of different individuals with CP to gain control of an online BCI system. The results indicate that six users could control a sensorimotor rhythm BCI and three a steady state visual evoked potential BCI at statistically significant levels of accuracy (SMR accuracies; mean ± STD, 0.821 ± 0.116, SSVEP accuracies; 0.422 ± 0.069).
Brain-computer interfaces (BCIs) have been proposed as a potential assistive device for individuals with cerebral palsy (CP) to assist with their communication needs. However, it is unclear how well-suited BCIs are to individuals with CP. Therefore, this study aims to investigate to what extent these users are able to gain control of BCIs.
METHODS:
This study is conducted with 14 individuals with CP attempting to control two standard online BCIs (1) based upon sensorimotor rhythm modulations, and (2) based upon steady state visual evoked potentials.
RESULTS:
Of the 14 users, 8 are able to use one or other of the BCIs, online, with a statistically significant level of accuracy, without prior training. Classification results are driven by neurophysiological activity and not seen to correlate with occurrences of artifacts. However, many of these users' accuracies, while statistically significant, would require either more training or more advanced methods before practical BCI control would be possible.
CONCLUSIONS:
The results indicate that BCIs may be controlled by individuals with CP but that many issues need to be overcome before practical application use may be achieved.
SIGNIFICANCE:
This is the first study to assess the ability of a large group of different individuals with CP to gain control of an online BCI system. The results indicate that six users could control a sensorimotor rhythm BCI and three a steady state visual evoked potential BCI at statistically significant levels of accuracy (SMR accuracies; mean ± STD, 0.821 ± 0.116, SSVEP accuracies; 0.422 ± 0.069).
2012
Daly, Ian; Grissmann, Sebastian; Brunner, Clemens; Allison, Brendan Z.; Müller-Putz, Gernot
Hybrid BCI classification via dynamic re-weighting Conference
Proceedings of the 3rd TOBI workshop, Würzburg, Germany, 2012.
Abstract | BibTeX | Tags: Classifier, EEG, ERD, Hybrid BCI, SSVEP
@conference{Daly2012,
title = {Hybrid BCI classification via dynamic re-weighting},
author = {Ian Daly and Sebastian Grissmann and Clemens Brunner and Brendan Z. Allison and Gernot Müller-Putz},
year = {2012},
date = {2012-10-01},
booktitle = {Proceedings of the 3rd TOBI workshop, Würzburg, Germany},
abstract = {A hybrid brain-computer interface (hBCI) may combine two or more BCI paradigms with the objective of improving performance (accuracy, stability, bit rate etc.) over that achievable with a single paradigm. However, the approach taken in some recent hBCI studies did not achieve accuracies significantly better than a single paradigm. Therefore, we introduce a re-weighting method for classifying a hybrid feature set. This approach produces higher accuracies than with the ERD paradigm.},
keywords = {Classifier, EEG, ERD, Hybrid BCI, SSVEP},
pubstate = {published},
tppubtype = {conference}
}
A hybrid brain-computer interface (hBCI) may combine two or more BCI paradigms with the objective of improving performance (accuracy, stability, bit rate etc.) over that achievable with a single paradigm. However, the approach taken in some recent hBCI studies did not achieve accuracies significantly better than a single paradigm. Therefore, we introduce a re-weighting method for classifying a hybrid feature set. This approach produces higher accuracies than with the ERD paradigm.