Back to the main page.

Bug 2880 - Identify bad channels with ft_rejectvisual without removing them

Reported 2015-04-16 16:03:00 +0200
Modified 2019-08-10 12:30:41 +0200
Product: FieldTrip
Component: preproc
Version: unspecified
Hardware: PC
Operating System: Linux
Importance: P5 normal
Assigned to:
Depends on:
See also:

- 2015-04-16 16:03:48 +0200

When dealing with 306-channel Neuromag system, we have to separately perform visual rejection for magnetometers and gradiometers. To correctly do so, I want to use cfg.keepchannel = 'yes' in order to separately identify bad magnetometers and bad gradiometers. Then I whish to be able to put all these bas channels together and remove them at once from the data. Nevertheless, I did not manage to find the information about the bad channels I selected in the output structure of the ft_rejectvisual function. Is there any way to have this information? Best. N.B. I also use cfg.keeptrial = 'yes' and manage to get bad trials indices without removing them thanks to the sample information in results.cfg.artfctdef.summary.artifact. Thanks to that information I was able to remove them at once contrary to the channels.

Robert Oostenveld - 2015-04-16 16:44:58 +0200

Hi Maxime, There is an example foe neuromag on But explains it better (at 2/3 on the page there is a box which specifically deals with this). With cfg.keepchannel=yes, the output data.cfg still will contain the channels that were selected (i.e. not deselected). Using union, intersect and the likes you can also make your own version.

Robert Oostenveld - 2015-11-10 16:28:46 +0100

cleaning time. I don't think that any code is needed. Better documentation would be useful, but I won't be writing that any time soon.

Robert Oostenveld - 2019-08-10 12:30:41 +0200

This closes a whole series of bugs that have been resolved (either FIXED/WONTFIX/INVALID) for quite some time. If you disagree, please file a new issue on