Back to the main page.
Bug 243 - ft_headmodel_halfspace should be implemented for EEG
| Status | CLOSED FIXED |
| Reported | 2010-11-27 13:24:00 +0100 |
| Modified | 2011-02-23 13:51:09 +0100 |
| Product: | FieldTrip |
| Component: | forward |
| Version: | unspecified |
| Hardware: | PC |
| Operating System: | Mac OS |
| Importance: | P1 normal |
| Assigned to: | Cristiano Micheli |
| URL: | |
| Tags: | |
| Depends on: | |
| Blocks: | |
| See also: |
Robert Oostenveld - 2010-11-27 13:24:19 +0100
this involves not only the function itself, but also prepare_vol_sens (projection of electrodes) and compute_leadfield
Robert Oostenveld - 2010-12-01 16:23:07 +0100
done, next is to implement it in ft_prepare_vol_sens and ft_compute_leadfield
Cristiano Micheli - 2011-01-28 15:25:55 +0100
(In reply to comment #1) > done, next is to implement it in ft_prepare_vol_sens and ft_compute_leadfield done, electrodes have to be projected onto the plane inly in the case of ECoG grid. For CSD of shaft electrodes is not necessary, so for the moment, for generality sake, no projection is implemented
Robert Oostenveld - 2011-01-28 15:33:21 +0100
projection should be done if the electrodes are on the "air side" of the plane. ft_headmodel_halfspace shoule be rewritten such that a plane is fitted to the geometry (1st input argument). The geometry only has to describe surface points, if it also contains tri, then that is not used (but should not case an error). furthermore, a test function should be implemented that compares a 100 meter single-sphere with a halfspace. They should be almost the same.
Cristiano Micheli - 2011-02-02 12:45:41 +0100
(In reply to comment #3) > projection should be done if the electrodes are on the "air side" of the plane. > done > ft_headmodel_halfspace shoule be rewritten such that a plane is fitted to the > geometry (1st input argument). The geometry only has to describe surface > points, if it also contains tri, then that is not used (but should not case an > error). > done > furthermore, a test function should be implemented that compares a 100 meter > single-sphere with a halfspace. They should be almost the same. done, but the comparison is between infinite and halfspace solution in case the dipoles are near/far away from the plane. To implement the sphere solution a mirror charge sphere solution should be added and tested against the halfspace, since the existing solution for conductive spheres assumes the potential to be piecewise-constant inside the concentric spheres.