Back to the main page.
Bug 373 - new function related to source space: ft_reducepatch
| Status | CLOSED WONTFIX |
| Reported | 2011-01-06 18:05:00 +0100 |
| Modified | 2011-01-26 15:31:55 +0100 |
| Product: | FieldTrip |
| Component: | forward |
| Version: | unspecified |
| Hardware: | PC |
| Operating System: | Windows |
| Importance: | P1 normal |
| Assigned to: | Jan-Mathijs Schoffelen |
| URL: | |
| Tags: | |
| Depends on: | |
| Blocks: | |
| See also: |
Lilla Magyari - 2011-01-06 18:05:44 +0100
There should be a new (FieldTrip compatible) function for reducing the number of points in a cortical mesh (source) (resulting in similar triangle sizes). This is for the development of minimum-norm estimate in FieldTrip. http://fieldtrip.fcdonders.nl/development/minimum_norm_estimate_in_fieldtrip?&#procedure http://fieldtrip.fcdonders.nl/development/replicate_functionality_of_mne_software?&#anatomical_processing Lilla
Robert Oostenveld - 2011-01-07 09:26:21 +0100
The development version of SPM8 includes functions for mesh handling. Could you have a look at those and judge whether they are of any use? ls spm8/spm_mesh* spm_mesh_adjacency.m spm_mesh_clusters.m spm_mesh_curvature.m spm_mesh_distmtx.m spm_mesh_edges.m spm_mesh_geodesic.m spm_mesh_get_lm.m spm_mesh_inflate.m spm_mesh_label.m spm_mesh_neighbours.m spm_mesh_normals.m spm_mesh_project.m spm_mesh_render.m spm_mesh_smooth.m spm_mesh_split.m spm_mesh_to_grid.m spm_mesh_transform.m spm_mesh_utils.m
Lilla Magyari - 2011-01-10 12:26:37 +0100
(In reply to comment #1) > The development version of SPM8 includes functions for mesh handling. Could you > have a look at those and judge whether they are of any use? > > ls spm8/spm_mesh* > > spm_mesh_adjacency.m > spm_mesh_clusters.m > spm_mesh_curvature.m > spm_mesh_distmtx.m > spm_mesh_edges.m > spm_mesh_geodesic.m > spm_mesh_get_lm.m > spm_mesh_inflate.m > spm_mesh_label.m > spm_mesh_neighbours.m > spm_mesh_normals.m > spm_mesh_project.m > spm_mesh_render.m > spm_mesh_smooth.m > spm_mesh_split.m > spm_mesh_to_grid.m > spm_mesh_transform.m > spm_mesh_utils.m I have looked at them. There is a function that is not listed here but maybe useful: spm_mesh_area that computes the surface area of a triangle mesh. It can return the whole area but also per faces. All the others are not dealing with reduction of points. I haven't really understood what spm_mesh_normals does. Lilla
Jan-Mathijs Schoffelen - 2011-01-10 15:01:21 +0100
Created attachment 15 figure
Lilla Magyari - 2011-01-10 15:27:01 +0100
(In reply to comment #1) there is also an mne matlab function: mne_reduce_surface but it is using also the matlab reducepatch function. However, here you have to give a ratio for the reduction and not the desired number of points. But in MNE when you setup the source space there is an option --ico. This option is reducing the source space in a way that it is using the topology of the recursively subdivided icosahedron. As a result, you can't reduced the points to any number but only to a set of numbers. (it is on page 22 and on page 98 in the tutorial). (But may you know this already.) Lilla
Jan-Mathijs Schoffelen - 2011-01-10 16:01:09 +0100
Yes, I saw the matlab function mne_reduce_surface as well. I don't see a problem with the --ico option; I took --ico 4, leading to 2562 points per hemisphere, which is a decent number.
Lilla Magyari - 2011-01-10 16:05:32 +0100
(In reply to comment #5) > Yes, I saw the matlab function mne_reduce_surface as well. > > I don't see a problem with the --ico option; I took --ico 4, leading to 2562 > points per hemisphere, which is a decent number. There is no problem with it. I just have drew the attention to the fact that there is a difference between how mne does and how FieldTrip does the reduction.
Robert Oostenveld - 2011-01-12 11:57:05 +0100
*** Bug 222 has been marked as a duplicate of this bug. ***
Robert Oostenveld - 2011-01-12 11:57:53 +0100
(In reply to comment #7) > *** Bug 222 has been marked as a duplicate of this bug. *** the solution for bug 373 should be implemented throughout FT and fix 222