Paper Info
Title
A Novel Multi-class Brain-Computer Interface (BCI) Paradigm Based on Motor Imagery Sequential Coding (MISC) Protocol.
Abstract
In this study, we present a novel multi-class BCI paradigm based on motor imagery sequential coding (MISC) protocol, which can generate multiple commands just by two kinds of motor imagery (MI) tasks. In the MISC protocol, each mental task was divided into several continuous epochs with the same duration. During each epoch, one of the two MI tasks was executed. With this protocol, multiple mental states can be coded by the two MI tasks. Additionally, the difficulty of classifier design was also reduced as only two MI tasks were needed to be classified. Three subjects participated in our experiments, and achieved an average accuracy of 85.7%, with the ITR of 16.5 bits/min. The results confirmed that the MISC protocol can generate more commands in BCI system with the equal number of MI tasks. © 2013 Springer-Verlag Berlin Heidelberg.
Year
DOI
Venue
2013
10.1007/978-3-642-42057-3-38
IScIDE
Keywords
Field
DocType
electroencephalogram (eeg),motor imagery sequential coding,multi-class brain-computer interface,sensorimotor rhythms
Computer science,Brain–computer interface,Speech recognition,Artificial intelligence,Classifier (linguistics),Sequential coding,Motor imagery
Conference
Volume
Issue
ISSN
8261 LNCS
null
16113349
Citations 
PageRank 
References 
0
0.34
6
Authors
6
Name
Order
Citations
PageRank
Jun Jiang1105.10
Erwei Yin21109.12
Yang Yu3385.71
Jingsheng Tang493.33
Zongtan Zhou541233.89
Dewen Hu61290101.20