Hello all--
I've been using parallelRecognition.py to repeatedly label a brain multiple times. I end up with a number of graphs of of the filename res_[0-9].arg and their corresponding .nrj files. However, I am unable to successfully open them in anatomist.
For example, attempting to open /home/dave/brainvisa/MRI/8087/graphe/res_0.arg via the Show Fold Graph dialog results in a brain mesh but nothing else. Is there something missing in the .arg files generated this way that prevents them from being viewed? Is there any way I can edit them such that they can be viewed? Thanks in advance.
-David
parallel recognition brains
-
riviere
- Site Admin
- Posts: 1361
- Joined: Tue Jan 06, 2004 12:21 pm
- Location: CEA NeuroSpin, Saint Aubin, France
- Contact:
Hi David,
I think I have never finished to correctly setup this process... I have to check it. What I suspect is that the "res_*.arg" are not (or ambiguously) recognized by BrainVisa database, so the corresponding nomenclature and colors are not loaded into Anatomist by the "Show folds arg" viewer.
You can do it manually from Anatomist:
- when the .arg is loaded in Anatomist, if the nomenclature "sulcal_roots_colors.hie" is not loaded, load it (in Anatomist): it's in the first preselected directory of the file browser.
- put it (the .hie object) in a browser window
- the .arg should be already in a 3D window (even if you see nothing: it's normal)
- if the .hie has been loaded after the .arg, go to "settings/graph parameters" (menu in Anatomist main window) and set "use attribute" to the value "label" (reselect it even if it is already set to "label" to force its application).
- unfold the nomenclature (or hierarchy, .hie) in the browser window, you will see hemispheres, lobes, sulci names
- click on a name (for instance "brain"): sulci included in this structure name will appear in red (selected). "brain" and "unknown" are enough to highlight all objects
- on the 3D view, you can unselect sulci by using "unselect" from the right-click popup
Note: In BrainVisa, in Parallel Recognition, the best result (with lowest energy) should be kept as the output_graph parameter. If you have left the default value, this one should be OK when viewed by the eye icon. Is it OK ?
Denis
I think I have never finished to correctly setup this process... I have to check it. What I suspect is that the "res_*.arg" are not (or ambiguously) recognized by BrainVisa database, so the corresponding nomenclature and colors are not loaded into Anatomist by the "Show folds arg" viewer.
You can do it manually from Anatomist:
- when the .arg is loaded in Anatomist, if the nomenclature "sulcal_roots_colors.hie" is not loaded, load it (in Anatomist): it's in the first preselected directory of the file browser.
- put it (the .hie object) in a browser window
- the .arg should be already in a 3D window (even if you see nothing: it's normal)
- if the .hie has been loaded after the .arg, go to "settings/graph parameters" (menu in Anatomist main window) and set "use attribute" to the value "label" (reselect it even if it is already set to "label" to force its application).
- unfold the nomenclature (or hierarchy, .hie) in the browser window, you will see hemispheres, lobes, sulci names
- click on a name (for instance "brain"): sulci included in this structure name will appear in red (selected). "brain" and "unknown" are enough to highlight all objects
- on the 3D view, you can unselect sulci by using "unselect" from the right-click popup
Note: In BrainVisa, in Parallel Recognition, the best result (with lowest energy) should be kept as the output_graph parameter. If you have left the default value, this one should be OK when viewed by the eye icon. Is it OK ?
Denis