Hi
I have a feeling that your registration would work if you used fslreorient on your structural scans.
Danny
________________________________________
From: FSL - FMRIB's Software Library [[log in to unmask]] on behalf of Dina Dajani [[log in to unmask]]
Sent: Friday, September 22, 2017 11:52 AM
To: [log in to unmask]
Subject: [FSL] FNIRT registration failure
Hello,
I am using FLIRT and FNIRT to register my functional data to MNI space. Registration works well between the functional and structural image (within subject), but fails during both the structural --> standard and functional --> standard steps (see attached photo). The brain extraction is good, the T1 image is fairly good (some ringing), the white matter segmentation is good. I do not have a b0 field map. I used linear registration with the BETted image as input and 6 DOF (I also tried BBR- same poor result); I used non linear registration with 12 DOF, 10 mm warp resolution. Both included "normal search". Any idea on how to improve the quality of the registration? It previously worked to normalize using EPI templates instead of the structural... is there a way to do this in FSL?
Thank you,
Dina Dajani
|