Performance analysis of a P300 BCI speller through single channel ICA
Christopher J. James, Sheng Wang · 2008
In this work we test a technique based on independent component analysis (ICA), applied to single channel brain signals recorded through the electroencephalogram. Standard (or ensemble) ICA (enICA) requires multiple channel recordings to work, however when single of few channels are required enICA cannot be readily applied. Single channel ICA (scICA) can be performed by using the method of delays we have previously proposed. Traditional source selection for scICA is to subjectively select related components based on prior knowledge. Here we trial an automatic source extraction method, which can increase the selection process speed and form an automatic robust system. These techniques are demonstrated on the P300 evoked potentials of a brain-computer interfacing (BCI) word speller dataset. Due to the poor SNR, as well as the presence of other artifacts, it is difficult to detect the P300 pattern on raw signal data. The results show that proposed algorithms are able to accurately and repeatedly extract the relevant information buried within noisy signals. These advantages are paramount for building a more reliable and robust system for use in real-world BCI - i.e. for use outside of the clinical laboratory.