segmentation from masks generated with other tools

[elabyt]

Dear all,
I try to do a segmentation of a pre-surgery MRI (with stereotactic frame). So, MRI contains artefacts from the stereotactic frame. I built head and brain masks with another tool and I would like to use these maske to build the corrersponding meshes in BrainVisa. The head and brain masks are in nifti format.
I tried to use the component "head mesh". I used folder icon to open my mask file and I loaded the corrected mri. But it seems that mesh file is also required. I don't well understand because it is the file that I want to create !!!
Could you help me ? How can I use my masks to do the corresponding meshes ?
Thanks
Etienne

[Dominique Geffroy]

Hi Etienne,

In HeadMesh process you can't give the head mask as an input. It is an output, it is computed in the process before computing the head mesh. I don't think it is possible to compute the head mesh from your head mask with this process. You may use directly the command AimsMeshBrain.
However, I think you could use the brain mask you computed with another tool. You can import it in Brainvisa with the process T1 MRI -> import -> Import Brain mask. It is the same procedure as importation of T1 MRI. You enter the brain mask as input with the folder icon, then you select the output attributes with the red database icon (database, protocol, subject...).
Then you can run the pipeline without step "brain mask segmentation" (unselect it to avoid erasing the mask you've just imported).

Dominique

[elabyt]

Hi Dominique,
I tried that you recommended me but when i use the AimsmeshBrain command, I have an error message:
what ( ) : bet_outskull_mask.nii: format error

which type of format is required for this command as input ?

I also tried to import my mask as you recommended and in this case, i also have an error message concerning my file:
invalid syntax (<string>, line 1)
But I can view this mask without problem with Anatomist. So, my file doesn't seem to be corrupted.
Thanks for your help
Etienne

[riviere]

Hi Etienne,

Nifti should be OK for AimsMeshBrain, actually the IO system is the same in all Aims commands and in Anatomist, so I cannot explain why it fails on your data. Could you tell us what was exactly the command and arguments that you used ?
You should use it like this:

Code: Select all

AimsMeshBrain -i bet_outskull_mask.nii -o head.mesh

Oh, or maybe it is because your input image voxel type is not signed short ? AimsMeshBrain only accepts signed short images. To convert it, use:

Code: Select all

AimsFileConvert -i bet_outskull_mask.nii -o bet_outskull_mask_S16.nii -t S16

Denis