Title
Enhancing performance of P300-Speller under mental workload by incorporating dual-task data during classifier training.
Abstract
•Performance of P300-Speller always impaired once applied in practical situations due to the effect of mental workload. We aimed to mitigate this effect and enhance its performance.•We propose a new method of incorporating dual-task data during classifier training and compare its performance with speller-only data training condition. Two task types were studied.•The performances of P300-Speller including character recognition accuracies and round accuracies under dual-task conditions were significantly improved. Further analysis of ERPs supported the results.•The findings in this study confirmed the feasibility of building a universal training model which can significantly mitigate the effects of mental workload on P300-Speller in its practical applications.
Year
DOI
Venue
2017
10.1016/j.cmpb.2017.09.002
Computer Methods and Programs in Biomedicine
Keywords
Field
DocType
Brain-computer interface,P300-Speller,Event-related potential,Mental workload,Training model
Training set,Character recognition,Workload,Computer science,Support vector machine,Event-related potential,Brain–computer interface,Speech recognition,Test data,Artificial intelligence,Classifier (linguistics),Machine learning
Journal
Volume
Issue
ISSN
152
C
0169-2607
Citations 
PageRank 
References 
0
0.34
7
Authors
10
Name
Order
Citations
PageRank
Yuqian Chen100.34
yufeng ke217.78
Guifang Meng300.34
Jin Jiang402.37
Hongzhi Qi54920.61
Xuejun Jiao602.03
Minpeng Xu72717.17
Peng Zhou8136.25
Feng He9169.45
Dong Ming1010551.47