On 12 Mar 2010, at 14:02, Matt Glasser wrote:
> If you really want to use rotated bvecs with two averages, you would be
> better off just stopping at the merge stage and merging the bvecs. The only
> downside to this is that stuff like BEDPOSTx will take longer.
>
> Peace,
>
> Matt.
>
> -----Original Message-----
> From: FSL - FMRIB's Software Library [mailto:
[log in to unmask]] On Behalf
> Of Andreas Bartsch
> Sent: Friday, March 12, 2010 12:07 AM
> To:
[log in to unmask]
> Subject: [FSL] AW: [FSL] rotating bvecs
>
> Well, if you do that you may as well use the unrotated bvecs, no?
> Cheers-
> Andreas
>
>
> ________________________________
> Von: FSL - FMRIB's Software Library [
[log in to unmask]] im Auftrag von
> wlzhu [
[log in to unmask]]
> Gesendet: Freitag, 12. März 2010 04:18
> An:
[log in to unmask]
> Betreff: Re: [FSL] rotating bvecs
>
> I have modified it to fit for file name is absolute path in *.ecclog.
>
> For use, we have two acquisition each subject and each has the same bvecs.
> We first merge the second image to the first one. Then perform eddy correct
> with reference of the b0 in first image.
> Next split the corrected image to 2 images, finally average them.
>
> The question is ,although two images have same bvecs, after using
> rotate_bves, the two bves are not same(slightly different).
> Is it acceptable to use the first rotated bvecs as the gradient of averaged
> image.
>
>
> thanks
>
>
> wanlin
>
> On Fri, Mar 12, 2010 at 11:18 AM, Mark Jenkinson
> <
[log in to unmask]<mailto:
[log in to unmask]>> wrote:
> Dear Gad,
>
> You might want to just use the existing fix described in the
> mailing list. This bug was noted and a fix posted which just
> involved changing one line. I've attached the fixed version.
>
> Note that this version behaves like the old one and keeps
> the naming of the bvecs file consistent so that it will work
> with other FSL tools such as bedpostx.
>
> All the best,
> Mark
>
>
>
>
>
>
>
> On 11 Mar 2010, at 23:23, Dave Deriso wrote:
>
> Hi Gad,
>
> I had the same promlem, so I just rewrote the script. It works pretty well
> for me. Give this a whirl and please be sure to let me know if it works
> without error!
>
> **Make sure you are in the same directory as the files when you run it.
>
> Attached is a zip file that contains:
>
> 1. rewritten script "rotate_bvecs2"
> 2. sample data "test.bvec" and "test.ecclog" these work reliably well for me
> 3. a sample output file "test_rot.bvec"
> 4. a text file with what the terminal should print
>
> Good luck!
>
> Best,
>
> Dave Deriso
> UCSD Psychiatry
>
>
> On Thu, Mar 11, 2010 at 2:04 PM, Gad Klein
> <
[log in to unmask]<mailto:
[log in to unmask]>> wrote:
> I am following the instructions on the FSL site to pre-process diffusion
> data prior to bedpostX. After eddy current correction worked nicely I tried
> to use the provided script to rotate by bvecs, but kept getting an error
> message. On the site the script is written as:
>
> rotate_bvecs <4doutput>.ecclog <bvecs>
>
> so I entered: rotate_bvecs
> /home/gklein/probDTI/DiffData/ambersleydata/DTI_Sat33+5/data.ecclog
> /home/gklein/probDTI/DiffData/ambersleydata/DTI_Sat33+5/RESUlug4s010a1001.bv
> ec
>
> and got this error message: [: 9:
> /home/gklein/probDTI/DiffData/ambersleydata/DTI_Sat33+5/RESUlug4s010a1001:
> unexpected operator
> /usr/share/fsl/bin/rotate_bvecs: 21: Syntax error: "done" unexpected
> (expecting "fi")
>
> Am I entering the bvec file incorrectly or something. Any help would be
> appreciated. Thanks.
>
> -Gad
>
> <test_rotate_bvecs.zip>
>