[Eeglablist] ICA (revised)

[nagwa ibrahim]

Dear sir, i'm working on removing artifacts from EEG signals using ICA as a pre-processing stage before feature extraction and classification stages. i use runica.m from eeglab toolbox, EEG signals are recorded from 7 channels, so i prepared data passed to runica.m function such that all epochs are concatenated thus data will be 7 chs* (epoch*samples/epoch). i read in documentation that i should use rmbase.m before passing data to runica.m, but i don't know why, so could you plesae guid me what are essential steps (or functions) tha should be used so that i do ICA in right way?? (as i'm currently used runica.m only without doing any other processing using any other functions in eeglab toolbox). also, i need to know can ICA improve classification results by only 1-2% as i see it so small improvment??? Thanks in Advance,Nagwa
_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today it's FREE!
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20080312/d4962891/attachment.html>