Back to the main page.

Bug 3025 - ft_prepare_sourcemodel: failure with resolution option without sensor description

Reported 2015-12-11 13:07:00 +0100
Modified 2016-01-14 21:25:59 +0100
Product: FieldTrip
Component: core
Version: unspecified
Hardware: PC
Operating System: Windows
Importance: P5 normal
Assigned to: Jan-Mathijs Schoffelen
Depends on:
See also:

Jan-Mathijs Schoffelen - 2015-12-11 13:07:23 +0100

This is hard-coded in the function. In principle it's OK, but I would suggest to relax this a bit, and allow the function to run through if a headmodel is specified (rather than a sens). As I see it, a geometry is needed to define the grid boundary. This shouldn't necessarily be the sensor positions, but could be the headmodel (after some outward shift. This came up when somebody was working with the (under construction) sourcemodel tutorial.

Jan-Mathijs Schoffelen - 2015-12-11 13:17:56 +0100

[jansch@mentat001 fieldtrip]$ svn commit -m "enhancement - also allow for a headmodel for automatic grid boundary detection when specifying resolution only" ft_prepare_sourcemodel.m Sending ft_prepare_sourcemodel.m Transmitting file data . Committed revision 10996.

Robert Oostenveld - 2015-12-11 13:26:27 +0100

(In reply to Jan-Mathijs Schoffelen from comment #0) the sensor array based inside detection was used for MEG local spheres, where the set of spheres itself was not appropriate for inside/outside detection.

Jan-Mathijs Schoffelen - 2015-12-11 13:37:04 +0100

Ah, there's the dinosaur in the closet... Right now, my change will lead to a crash if someone puts a T-rex in the cfg, before it wouldn't have worked either, so nothing lost yet. I'll update.