Back to the main page.
Bug 2249 - ft_volumesegment does not return .anatomy field anymore
| Status | CLOSED WONTFIX |
| Reported | 2013-08-13 10:07:00 +0200 |
| Modified | 2014-01-29 13:28:32 +0100 |
| Product: | FieldTrip |
| Component: | core |
| Version: | unspecified |
| Hardware: | PC |
| Operating System: | Windows |
| Importance: | P3 normal |
| Assigned to: | Eelke Spaak |
| URL: | |
| Tags: | |
| Depends on: | |
| Blocks: | |
| See also: |
Eelke Spaak - 2013-08-13 10:07:28 +0200
Given an mri structure with a .anatomy field, ft_volumesegment used to return a structure with the masks (.white, .gray, .csf) *and* .anatomy, where .anatomy was (I assume) simply the same as in the original non-segmented mri. Today I noticed that the .anatomy field is missing from ft_volumesegment output. This messses up ft_prepare_sourcemodel: Error using ft_volumenormalise (line 109) no anatomical information available, this is required for normalisation Error in ft_prepare_sourcemodel (line 647) normalise = ft_volumenormalise(tmpcfg,mri);
Jan-Mathijs Schoffelen - 2013-08-13 10:46:49 +0200
As far as I know ft_volumesegment has never returned an anatomy field. It seems you want to create a source model based on the inverse warp of a template grid. To this end, you need to input the normal anatomical image into ft_prepare_sourcemodel The segmentation is needed for ft_prepare_headmodel
Eelke Spaak - 2013-08-13 10:52:55 +0200
(In reply to comment #1) That's true. But I'm pretty sure that in the past I just put in the segmented volume into both ft_prepare_headmodel and ft_prepare_sourcemodel and got sensible results. (I just checked my previously saved segmented mri structures, and they indeed contained a .anatomy.) But maybe we should just close this one then?
Jan-Mathijs Schoffelen - 2013-08-14 13:32:26 +0200
.anatomy should be there, problem might have arisen by also allowing FSL segmentation.
Jan-Mathijs Schoffelen - 2013-08-14 13:33:58 +0200
or, more likely: we used to copy the input structure and return a modified version, now the output structure is created from scratch, with individual fields being explicitly copied (if applicable).
Jan-Mathijs Schoffelen - 2013-08-14 13:35:27 +0200
JM thinks I'm still wrong (me=Eelke). check that ft_datatype return segment/volume properly, even if .anatomy is present in segmentation.
Eelke Spaak - 2013-10-22 11:57:43 +0200
This bug is not a bug, and also not important enough to care about.