Hi Adam,
the long stream is designed to reduce processing noise across longitudinal data. On one hand this can help when comparing arcoss scanners or acquisition, as the methods will be more sensitive to the scanner/acquisition induced differences. However, when switching resolutions, this approach is dangerous. The subject template (base) needs to be in some space. If we chose .8 mm for the base, then surfaces may go deeper into the sulcii and they may stay there even for 1mm long runs. On the contrary, when choosing a 1mm base, surfaces may not go deep into some of the sulcii and the long pipeline (3rd stage, long) may not be able to push them there.
So for analysis across resolution I would recommend to only use the cross stream.
Best, Martin
On 05/02/2017 09:17 PM, Adam Martersteck wrote:
Just bumping to the top to see if anyone has ideas about comparing 0.8mm vs. 1mm data using the long stream.
Dear FS Experts,
I’m trying to compare multiple T1-weighted sequences taken across multiple scanners in the same day in 15 individuals using the longitudinal stream.
One question we had was how different cortical thickness estimates would be using a 0.8mm^3 resolution vs. our standard 1mm^3 MPRAGE.
Unfortunately, during the initial long-stream norm fusion, the –hires enabled 0.8mm data has a mismatch in dimensions compared to the standard 1mm data.
Given this limitation -- what would be the best way to compare these two sequences?
Thanks, Adam
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer