Back to the main page.

Bug 1135 - Remove neighbourtemplates for MEG systems that are based on 2D layouts?

Reported 2011-11-10 15:57:00 +0100
Modified 2014-03-12 12:21:35 +0100
Product: FieldTrip
Component: core
Version: unspecified
Hardware: All
Operating System: All
Importance: P3 enhancement
Assigned to: Jörn M. Horschig
Depends on:
See also:

Jörn M. Horschig - 2011-11-10 15:57:36 +0100

When creating neighbour templates, I included templates based on all 2D layouts as well. However, these neighbour definitions are not generalizable to real 3D gradiometer data. E.g. for the neuromag306 system, we now got several templates based on the 2D templates: neuromag306mag_neighb neuromag306grad_neighb neuromag306planar_neighb And one based on the test data (3D gradiometer information) neuromag306_neighb It might be confusing for some people that there are different templates for the same system. Also the question whether it makes sense at all to support 'wrong' neighbour templates can be asked Any thoughts on this?

Boris Reuderink - 2011-11-17 10:46:37 +0100

Changed the status of bugs without a specific owner to UNCONFIRMED. I'll try to replicate these bugs (potentially involving the submitter), and change confirmed bugs to NEW. Boris

Boris Reuderink - 2011-11-17 13:40:44 +0100

Hmm. I don't really know what to do with this bug report. Is this something to discuss in the meeting? Or on a FieldTrip mailing list?

Jörn M. Horschig - 2011-11-17 14:45:44 +0100

I don't know, this has to do MArco Buiatti's recent mails in the mailing list. You could read up on them and decide then? I am not sure, especially because we do not work with neuromag data here at the DCCN. Robert might know more about this, best would be to ask and discuss with him.

Boris Reuderink - 2011-11-18 15:07:44 +0100

Robert, could you think of an appropriate follow-up for this remark of Jörn?

Boris Reuderink - 2012-02-01 15:45:26 +0100

Robert explained. My understanding of the appropriate action: - Modify code to yield warning, point to FAQ that explains difficulties of deriving neighbors based on gradiometers & magnetometers. - Delete 2D neighbor templates. - Involve Marco Buiatti to elaborate new FAQ entry. Boris

Jörn M. Horschig - 2012-08-23 13:48:32 +0200

added crossreference to bug800 - I'm gonna check the 2D templates after having proper neighbour templates and decide upon the quality of the neighbour templates on 3D sensor positions whether the 2D templates are useful or not. In any case, I'll also work on this bug as decided (e.g. throw a warning)

Jörn M. Horschig - 2013-03-13 10:04:38 +0100

- added to ft_prepare_neighbours: if strcmp(ft_senstype(sens), 'neuromag306') warning('Neuromagr06 system detected - be aware of different sensor types, see'); end svn ci ft_prepare_neighbours.m -m "enhancement #1135 - pointing to different sensor types for neuromag system" Sending ft_prepare_neighbours.m Transmitting file data . Committed revision 7653. - thus also created - and modified neuromag neighbour templates accordingly: svn ci -m "bugfix #1135 - corrected for neuromag306 apartheid - mag only goes with mag, grad only with grad" Sending neighbours/neuromag306_neighb.mat Deleting neighbours/neuromag306cmb_neighb.mat Deleting neighbours/nm306all_neighb.mat Deleting neighbours/nm306mag_neighb.mat Deleting neighbours/nm306planar_neighb.mat Transmitting file data . Committed revision 7652. Note that the neuromag306_neighb.mat is the concatenation of the magnetometer and the gradiometer definition, i.e. magnetormeters and gradiometers are still separated, but all sensors are in one file now.

Jörn M. Horschig - 2013-03-13 10:05:37 +0100

Hi Robert, I fixed the neuromag306 neighbour issue and then realized you were not in the CC list. Please check the last comment (==fix) in this bug

Robert Oostenveld - 2013-03-13 10:13:21 +0100 states "This topic does not exist yet". Please document the templates that you do want to keep. Have a look at the documentation of the other templates as example.

Jörn M. Horschig - 2013-03-13 10:57:21 +0100

created I am not sure whether picture would help much, because of the templates are based on 3D positions (which are hard to grasp without an interactive viewmode - maybe we could create gifs?). And I still need to remove some templates from the directory let me know if things should be adjusted on the template page

Jörn M. Horschig - 2013-03-20 14:11:16 +0100

see bug 1929 for the documentation for this bug I still need to: - remove the concatenated neuromag neighbour structure - create a neighbour structure for the combine planar representation of the neuromag306 system

Jörn M. Horschig - 2013-03-20 14:55:03 +0100

svn ci -m "enhancement #1135 - re-added combined planar neighbour definition, deleted concatenated all neighbour template" Deleting neighbours/neuromag306_neighb.mat Adding (bin) neighbours/neuromag306cmb_neighb.mat Transmitting file data . Committed revision 7697.