External Email - Use Caution        

Dear Douglas,
Dear All,

Trying to investigate the shift, I used the same pair of commands to convert several mgz files back to NIFTI and overlay on the original input T1W image. Some sort of mismatch begins with the orig.mgz file.  I see small local compressions and stretches. Since this is the image that will be subsequently segmented, segmentation will not match input T1W image. Subsequent files: nu.mgz, T1.mgz and brainmask.mgz carry over distorted (compared to input) orig.mgz but appropriately modify intensity and masking as expected.

However, I do not see overall 1 pixel shift in y-axis compared to the input image. This leads me to believe MGZ to NIFTI conversion is not the cause of the shift per se because I assume orientation information inside aseg.mgz is the same as in the other files. This of course could be wrong as I do not know details.

Thank you,

Roman

From: freesurfer-bounces@nmr.mgh.harvard.edu <freesurfer-bounces@nmr.mgh.harvard.edu> on behalf of Roman Fleysher <roman.fleysher@einsteinmed.org>
Sent: Tuesday, February 23, 2021 10:54 AM
To: freesurfer@nmr.mgh.harvard.edu <freesurfer@nmr.mgh.harvard.edu>
Subject: Re: [Freesurfer] segmentation is shifted with respect to NIFTI T1W
 
CAUTION: This email comes from an external source; the attachments and/or links may compromise our secure environment. Do not open or click on suspicious emails. Please click on the “Phish Alert” button on the top right of the Outlook dashboard to report any suspicious emails.

        External Email - Use Caution        

Thank you, Douglas. Then, fsroi as a work around is also right solution? I do not know what causes the issue: initial conversion of T1W from NIFTI to mgz, segmentation itself or transfer back to NIFTI. I do not know if other files (say brain mask) have the same issue. Looking into these.

Roman
From: freesurfer-bounces@nmr.mgh.harvard.edu <freesurfer-bounces@nmr.mgh.harvard.edu> on behalf of Douglas N. Greve <dgreve@mgh.harvard.edu>
Sent: Tuesday, February 23, 2021 10:04 AM
To: freesurfer@nmr.mgh.harvard.edu <freesurfer@nmr.mgh.harvard.edu>
Subject: Re: [Freesurfer] segmentation is shifted with respect to NIFTI T1W
 
Those commands look right

On 2/23/2021 9:27 AM, Roman Fleysher wrote:

        External Email - Use Caution        

Dear FreeSurfers,

I am working with version 6, starting point is T1W image in NIFTI format. I typically transform segmentations back to native space in NIFTI format to be able to use other software (FSL and such). I noticed that segmentation is shifted by 1 pixel in one axis (y-axis). Mapping of segmentation is a two step process based on description in:

MailScanner has detected a possible fraud attempt from "secure-web.cisco.com" claiming to be http://surfer.nmr.mgh.harvard.edu/fswiki/FsAnat-to-NativeAnat


Specific commands are (schematically) :

# realign to original MGZ

mri_label2vol --seg inputSegmentationInMGZ --temp mri/orig/001.mgz --o outputSegmentationInOrigMGZ --regheader inputSegmentationInMGZ


# convert to NIFTI (*.nii.gz)

mri_convert --in_type mgz --out_type nii outputSegmentationInOrigMGZ outputSegmentationInOrigNIFTI.nii.gz

Are these steps correct? I can shift NIFTI by one pixel using fslroi, but I want to be sure nothing else is happening and that this shift is the correct work around.

Thank you,

Roman



_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
MailScanner has detected a possible fraud attempt from "secure-web.cisco.com" claiming to be https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer