Back to the main page.

Bug 2389 - 4D-BTi-EEG channels are not recognized with our data

Reported 2013-11-26 12:04:00 +0100
Modified 2014-02-24 10:56:33 +0100
Product: FieldTrip
Component: fileio
Version: unspecified
Hardware: PC
Operating System: Linux
Importance: P3 normal
Assigned to: Robert Oostenveld
Depends on: 2027
See also:

- 2013-11-26 12:04:55 +0100

Created attachment 564 changed version of ft_chantype.m see lines 345 and 346 We have a 4D-BTi 248-MEG/32-EEG-channel system. In our case EEG channel labels end with '-1' rather than begin with 'E' (as expected in ft_chantype.m). See lines 345 and 346 in the attached file, line 345 is the commented original code, and 346 is a proposed solution. Maybe there are other 4D-BTi machines (or the labels can be customized in the machine settings), hence I think that the best solution is to recognize both the labels starting with 'E' and those ending with '-1' as EEG channels.

Robert Oostenveld - 2013-11-26 12:27:04 +0100

thanks for the clear suggestion. Since it is within the elseif ft_senstype(input, 'bti') ... section there should not be any side effects for non-4D datasets. I have added it to the fieldtrip code, it will be in the release this night. mac001> svn commit fileio/ft_chantype.m Sending fileio/ft_chantype.m Transmitting file data . Committed revision 8855. PS could you add your system to

Jan-Mathijs Schoffelen - 2013-11-26 13:05:50 +0100

Hi Pawel, Thanks for reporting. Where's the machine located that you are talking about? The 4D systems are notorious for being inconsistent with channel names etc across sites. For this reason it is hard to provide foolproof and robust solutions in the reading routines. Therefore, we are trying to compile a set of test datafiles that we can use to optimize the code (see bug 2027 for more info). Could you provide us with such a small dataset. What I would need is just a small binary file (can be c,rfDC or e,rfhp...) + the corresponding config file. Your request is reasonable, but I would like to have some testdata to try it out myself. Also, we should probably consider not relying on the channel names, but use the information that is in the binary to classify the channels, i.e. if I recall correctly, each channel also has a number attached that designates its class.

- 2013-11-26 14:14:40 +0100

(In reply to comment #2) Thanks a lot, we are already on the map, greetings from Magdeburg - Leibniz Institute for Neurobiology :) The smallest dataset we currently have is about 70MB, is it ok? How can I send such a big file to you? (I need to ask the head of my lab if I can; we can prepare smaller file but in about 2 weeks because now MEG is under being maintained)

Robert Oostenveld - 2013-11-26 14:16:48 +0100

(In reply to comment #3) ah, I had been searching for a MEG system in Warsaw, based on your email address! 70MB is fine. See at the bottom of and send the email to me or Jan-Mathijs in person.

Robert Oostenveld - 2014-02-24 10:56:33 +0100

I closed several bugs at once that all have been resolved for some time. If you disagree, please reopen.