Dear all,
I'm processing native 250 µm T1-weighted MPRAGE data with v6. I downsampled the data gradually from 250 µm to 1 mm and am processing it with the -hires flag, openmp and expert options affecting mris_inflate and mris_make_surfaces.
After intensity normalization of autorecon1 the white matter intensity is not set around 110 as it should be, but varies across the brain. Please find attach a screenshot of nu.mgz prior to the normalization and of T1.mgz after normalization. This happened in case of 250 µm data only, please also find attached the recon-all.log
Having pre-processed the data slightly differently I have encountered the same with the 330 µm, but not in case of 250 µm. However, downsampling the nu.mz and running mri_normalize on the downsampled nu.mgz resulted in a usual T1.mgz in all cases tested.
I can upload the subject folder if you wish. Folder size is approximately 1.3 GB without rawavg.mgz and 001.mgz
Best, Falk
---------------------------------------- Falk Lüsebrink, M. Sc.
Otto-von-Guericke-University, Magdeburg, Germany Faculty of natural sciences Institute for experimental physics
Biomedical Magnetic Resonance (BMMR) http://www.bmmr.ovgu.de/mm/en/
sure, if you upload I'll take a look
cheers Bruce
On Tue, 23 May 2017, Falk Lüsebrink wrote:
Dear all,
I’m processing native 250 µm T1-weighted MPRAGE data with v6. I downsampled the data gradually from 250 µm to 1 mm and am processing it with the -hires flag, openmp and expert options affecting mris_inflate and mris_make_surfaces.
After intensity normalization of autorecon1 the white matter intensity is not set around 110 as it should be, but varies across the brain. Please find attach a screenshot of nu.mgz prior to the normalization and of T1.mgz after normalization. This happened in case of 250 µm data only, please also find attached the recon-all.log
Having pre-processed the data slightly differently I have encountered the same with the 330 µm, but not in case of 250 µm. However, downsampling the nu.mz and running mri_normalize on the downsampled nu.mgz resulted in a usual T1.mgz in all cases tested.
I can upload the subject folder if you wish. Folder size is approximately 1.3 GB without rawavg.mgz and 001.mgz
Best,
Falk
Falk Lüsebrink, M. Sc.
Otto-von-Guericke-University, Magdeburg, Germany
Faculty of natural sciences
Institute for experimental physics
Biomedical Magnetic Resonance (BMMR)
freesurfer@nmr.mgh.harvard.edu
-
Bruce Fischl -
Falk Lüsebrink