I am trying to run recon-all with a T1w image and a FLAIR image to improve the removal of the dura. I am using the command
recon-all -sd /cenc/mri/freesurfer -subjid 34P1001 -all -i /cenc/mri/subjects/34P1001/freesurfer/input/t1w.nii.gz -FLAIR /cenc/mri/subjects/34P1001/freesurfer/input/t2flair.nii.gz -FLAIRpial
I have two questions. First, in reading the forums I read several threads on the forum saying that you run the recon-all without the -FLAIRpial option and then run
recon-all -autorecon3 -s $SUBJECT -FLAIRpial — (http://www.mail-archive.com/freesurfer@nmr.mgh.harvard.edu/msg28033.html)
Is it necessary to run -autorecon3 with -FLAIRpial option after recon-all has been run with the T1w image? It seems that if the -FLAIRpial option only is needed in -autorecon3 then it should be able to be specified simultaneously. I did try it the way posted with similar results for some images.
My biggest concern is that there is a severe misregistration between my T1w image and my FLAIR image. What can I do to improve the registration? The raw data are almost perfectly aligned (i.e. the subject did not move between acquisitions) so this misregistration has me baffled.. After registration the FLAIR image has been shifted and rotated. There is also a hyper intensity shading on the superior brain regions (best seen in the sagittal view). I have posted the mri and scripts folder to Google Drive if anyone cares to take a look
https://drive.google.com/drive/folders/0B4bmIh8K5WVJbDRpNXh1TVEzdVU?usp=shar...
I have also included 3 screen shots showing the issue of misregistration and hyper intensities. Any suggestions on how to resolve this issue would be greatly appreciated.
Thanks for your help in advance
Bob