 |
 |
Hi Christian,
Thanks for the note – I have been using ‘applywarp’ instead of flirt so that I can include fieldmap unwarping. Do you think that’s still OK?
David
From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On Behalf Of Christian F. Beckmann
Sent: Monday, December 19, 2011 11:21 PM
To: [log in to unmask]
Subject: Re: [FSL] resample image data
Hi,
Note that for best registration accuracy you should register to your 1mm isotopic atlas, save the transform and them use flirt on the original data together with the applyisoxfm option to simultaneously transform and down sample - no need to create the atlas in 3x3x3 resolution, other than maybe having an underlying image for fslview or overlay.
Hth
Christian
On 20.12.2011, at 04:38, David Grayson <[log in to unmask]> wrote:
OK – problem solved – I found my answer on the nitrc forum
flirt -in atlas -ref atlas -o atlas_333 -applyisoxfm 3
sorry to take up space!
David
From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On Behalf Of David Grayson
Sent: Monday, December 19, 2011 7:16 PM
To: [log in to unmask]
Subject: [FSL] resample image data
Hi FSL users,
I am wondering how to resample an image by any arbitrary factor, and keep the same centering? fslmaths -subsamp2 only allows for a factor of 2.
For instance, I’d like to resample my high-res atlas image from 1x1x1 mm to 3x3x3 mm, so that when I register my filtered_func_data to my atlas, I can use my atlas_333 image as reference, and my func data will be aligned and in 333 space.
Thanks very much for any insight,
David