Hi Douglas,

The brainmask volumes are extremely grainy (similar to the raw structurals), i.e., dark voxels throughout the WM and high-intensity voxels across the gray matter. As far as I understood, all WM voxels (unless a lesion is present) should have a uniform intensity distribution after the normalization stages within the volume processing pipeline. Moreover, this makes the WM edits particularly challenging, for it's very difficult to trace the boundaries of some of the WM strands.

I don't know if this represents an issue but it's definitely something I hadn't encountered before. Any input would be really appreciated!

Thanks,

--G

On Jun 19, 2013, at 2:23 AM, Douglas Greve <greve@nmr.mgh.harvard.edu> wrote:

Hi Gabriel, I don't have experience with the MR-PET yet. Why do you think there is a problem? Try running wm-anat-snr on this subject. In general, you don't want to be smoothing/denoising the anatomical
doug





On 6/18/13 1:13 PM, obregon@nmr.mgh.harvard.edu wrote:
Hi,

I've started segmenting several structural scans from an MR-PET study
conducted in Bay 7, and was wondering if someone has previous experience
working with MPRAGE images acquired using the MR-PET coils?

The scans have a very low SNR and I am trying to figure out if this could
potentially affect the performance of the automated surface
reconstructions. I've scrolled through the volumes and even though the
surfaces seem normal, I am somewhat skeptical about their true quality
given that intermediate stages, such as the intensity normalization, seem
to be failing (please find an example from one of the brain masks
attached).

Is it essential to filter the raw structurals before segmenting them, and
if so, are there any readily available denoising tools? Furthermore, are
there any additional issues I should take into consideration?

Thanks for your help!

Gabriel


_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer