[Eeglablist] BrainVision files

[Quentin Noirhomme]

Dear Lars,

The Fasst toolbox has been developed for EEG/fMRI correction for Brain 
vision files (other formats accepted too). It's based on spm8 but while 
the file are corrected and downsampled you can export it to EEGlab (this 
is what spm developpers say).

http://www.montefiore.ulg.ac.be/~phillips/FAST.html

It works in our lab with 5000 kHz sampling rate and long lasting study.

Best

Quentin



Gregor Volberg a écrit :
> Dear Lars,
> 
> I cannot answer your second question, but I had similar problems with reading 
> large files (~ 2.5 GB, 5000Hz sampling rate) from combined EEG/fMRI 
> recordings. You might  get it working with the following tricks:
> 
> (1) I moved from a 32-bit windows to 64-bit Linux and added RAM (8 GB; 16 GB 
> would be ideal but the mainboards of our standard Dell computers do not 
> support this). I also tried out the performance on the same hardware with a 
> dual boot (Windows/Linux) setup and found that the Linux system produced 
> generally less out-of-memory errors. 
> 
> (2) Even with the 8GB RAM/64bit Linux I could not read in the whole 2.5 GB 
> *eeg file. I used the pop_loadbv function to read in slices of of data 
> (corresponding to our experimental blocks), performed the gradient artifact 
> correction, and then downsampled and re-merged the slices. If the are 
> reasonable cut-points in your data, e. g. pauses between blocks, you might 
> use a similar strategy. Cut-points might be identified by reading in only the 
> header file, without data, and then look for an adequate sample point range. 
> 
> info = read_event('eh_w_22_l.vmrk') % reads in an event structure
> % then identify reasonable sample 
> 
> EEG = pop_loadbv(path, file, [start1 stop1]); %read in slice
> EEG = pop_fmrib_fastr(EEG,0,4,21,'R128',1,0,0,0,0,0.0125,[],'auto');
> EEG = pop_resample(EEG, 500);
> [EEG ALLEEG] = pop_saveset(EEG, 'filename', 'slice1.set', 'filepath', 
> filepath);
> 
> Finally, merge the slices with pop_mergeset (or via the eeglab GUI).
> 
> Hope that helps,
> Gregor
> 
> 
> Am Mittwoch, 28. Oktober 2009 16:34 schrieb Michels Lars:
>> Dear EEGlab members,
>>
>>
>>
>> I am currently performing a MR gradient correction for simultaneously
>> recorded EEG-fMRI data in EEGlab.
>>
>>
>>
>> In principle it works, however, it works only for files with a sampling
>> rate well below 5000 Hz. Otherwise I get an error message: Out of memory. I
>> thought the correction will work best for files with the (original)
>> sampling rate of 5000 Hz. Is there any way to import the large raw file to
>> EEGlab (file size is about 600 MB)?
>>
>>
>>
>> Second question: There is an option to export the file to BrainVision in
>> EEGlab. Does anybody now whether the (exported) file can be read/imported
>> in BrainAnalyzer?
>>
>>
>>
>> Best
>>
>> Lars
>>
>>
>>
>> Dr. phil. Lars Michels
>> Postdoctoral Fellow
>> Universitäts-Kinderspital
>> Steinwiesstrasse 75
>> 8032 Zürich
>> Tel.  0041 44 266 78 28
>> Fax  0041 44 266 71 53
>> http://www.kispi.uzh.ch/af/ForschungLehre/zentrum_de.html
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu


-- 
Quentin Noirhomme
Coma Science Group
http://www.comascience.org
Cyclotron Research Center
University of Liege
Allée du 6 Aout, 8
4000 Liège
+32(0)4 366 23 35