RT info:eu-repo/semantics/conferenceObject
T1 A Novel Hybrid Swarm Algorithm for P300-Based BCI Channel Selection
A1 Martínez Cagigal, Víctor
A1 Santamaría Vázquez, Eduardo
A1 Hornero Sánchez, Roberto
AB Channel selection procedures are essential to reduce thecurse of dimensionality in Brain-Computer Interfacesystems. However, these selection is not trivial, due tothe fact that there are 2Nc possible subsets for an Ncchannel cap. The aim of this study is to propose a novelmulti-objective hybrid algorithm to simultaneously: (i) reducethe required number of channels and (ii) increase theaccuracy of the system. The method, which integratesnovel concepts based on dedicated searching and deterministicinitialization, returns a set of pareto-optimalchannel sets. Tested with 4 healthy subjects, the resultsshow that the proposed algorithm is able to reach higheraccuracies (97.00%) than the classic MOPSO (96.60%),the common 8-channel set (95.25%) and the full set of 16channels (96.00%). Moreover, these accuracies have beenobtained using less number of channels, making theproposed method suitable for its application in BCIsystems.
YR 2018
FD 2018
LK http://uvadoc.uva.es/handle/10324/31358
UL http://uvadoc.uva.es/handle/10324/31358
LA eng
NO Producción Científica
DS UVaDOC
RD 22-dic-2024