Hi Antonin
yes, V6 was not meant for T2/FLAIR and we didn't test it extensively. We have been optimizing things in the past week for T2/FLAIR and I expect 6.1 will be better for them. We can give you a beta soon if you like.
cheers Bruce On Tue, 7 Feb 2017, Antonin Skoch wrote:
Dear experts,
I have reprocessed my groups of subjects (originally processed by v5.3) by v6.0. I am using -FLAIRpial for pial surface refinement.
I have noticed that the pial surfaces in v6.0 after -FLAIRpial are positioned much more externally in comparison to -FLAIRpial in v5.3. The v6.0 version includes to pial sheet much more partial volume voxels in comparison to v5.3.
In contrast to v5.3, where the surfaces after -FLAIRpial were mostly internally with respect to pial surfaces determined by T1, in v6.0 there are many regions where the pial surfaces after -FLAIRpial are placed much externally with respect to pial surfaces determined by T1.
See the screenshot (red surfaces are pial surfaces after -FLAIRpial with v6.0, yellow are surfaces after -FLAIRpial with v.5.3).
I have used -nsigma_above 8 in mris_make_surfaces to prevent excluding some cortical regions with high signal intensity but this I used consistently in both versions so it should not be the reason of difference.
I have uploaded the subject processed both by 5.3 and 6.0 version to your server, files FLAIRpial_v5.3.tar.gz and FLAIRpial_v6.0.tar.gz.
Could you please look at my data and check whether the pial surfaces after -FLAIRpial in v6.0 look OK from your perspective?
Regards,
Antonin Skoch