Print

Print
Dear Ana,

it doesn’t look like this has anything to do with the rotation in eddy. As I suspected those rotations are very small. It is instead a matter of the bvecs not being oriented correctly for FSL to begin with. 

It is important to get the axes of the bvec coordinates to be the same as the axes of the image volumes, and also to make sure that the signs are the same (for example that a larger number means further to the right) in  bot bvecs and the images. This is always hard to get right when starting to do diffusion with a new scanner and/or diffusion analysis software. What we recommend if one doesn’t know (https://fsl.fmrib.ox.ac.uk/fsl/fslwiki/FDT/UserGuide#Processing_pipeline) is to use trial and error until one sees the V1 vectors from a DTI analysis align across pixels. That is very easy to do in the brain where one “knows” how things should go and the directions are very distinct. Maybe it is harder to do in the pelvis?

In your case, since you say that the dsistudio bvecs work for tractography, it seems like sign-swapping the second column of the bvecs does the trick. At least it seems that is what dsistudio has done for you. Maybe someone with experience of dsistudio knows why it does that?

Jesper

On 24 May 2019, at 16:21, Ana Sofia Castro Verde <[log in to unmask]> wrote:

Dear Jesper,

I don’t think it is the case of the subjects being belly dancers ;)

Sure, I've added the bvecs .txt files as attachments (the original bvecs, and both after correction: using DSI Studio and eddy from FSL) so that you can take a look.

Thank you for your attention.

Best regards,

Ana Sofia

########################################################################

To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1


> No dia 24/05/2019, às 15:36, Jesper Andersson <[log in to unmask]> escreveu:
>
> Dear Ana,
>
> this sounds very strange. Typically the rotation is very small even for the head, and the rotated bvecs differ only very marginally from the original ones. I would have thought that, unless your subjects are all belly dancers, that rotation of the pelvic area is even smaller than for the head.
>
> Can you please send me the original and the eddy-rotated bvecs so I can see how much they have been rotated please?
>
> Jesper
>
>
>> On 24 May 2019, at 15:13, Ana Sofia Castro Verde <[log in to unmask]> wrote:
>>
>> Also, I would like to add a piece of information that might be helpful for those trying to solve this problem.
>>
>> When I use the bvecs table generated by eddy/motion correction from DSI Studio for the fiber tracking on DSI Studio using the output from eddy, the tracking seems to work fine.
>>
>> This makes me question why the rotated bvecs from eddy do not work on my data.. Does someone know if perhaps the algorithm behind the rotation of the bvecs on eddy has some assumption for the brain?
>>
>> Best regards,
>>
>> Ana
>>
>>> No dia 22/05/2019, às 20:10, Glasser, Matthew <[log in to unmask]> escreveu:
>>>
>>> Is that an isotropic resolution?
>>>
>>> Matt.
>>>
>>> On 5/22/19, 2:03 PM, "FSL - FMRIB's Software Library on behalf of Ana Castro Verde" <[log in to unmask] on behalf of [log in to unmask]> wrote:
>>>
>>>  Hi Matt,
>>>
>>>  Thanks for answering.
>>>
>>>  These are the acquisition parameters:
>>>  - resolution: 1.39 mm
>>>  - b-value: 600 s/mm^2
>>>  - number of directions: 16
>>>
>>>  Also, there is this paper that I've been using as main reference to my project, where they show the feasibility of performing fiber tracking of the male urethral sphincter complex:
>>>  S. Sinha, U. Sinha, V. Malis, V. Bhargava, K. Sakamoto, and M. Rajasekaran, “Exploration of male urethral sphincter complex using diffusion tensor imaging (DTI)-based fiber-tracking,” J. Magn. Reson. Imaging, pp. 1–10, 2018.
>>>
>>>  Let me know if you have any thoughts.
>>>
>>>  Best regards,
>>>
>>>  Ana
>>>
>>>  ########################################################################
>>>
>>>  To unsubscribe from the FSL list, click the following link:
>>>  https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
>>>
>>>
>>>
>>> ________________________________
>>> The materials in this message are private and may contain Protected Healthcare Information or other information of a sensitive nature. If you are not the intended recipient, be advised that any unauthorized use, disclosure, copying or the taking of any action in reliance on the contents of this information is strictly prohibited. If you have received this email in error, please immediately notify the sender via telephone or return mail.
>>>
>>> ########################################################################
>>>
>>> To unsubscribe from the FSL list, click the following link:
>>> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
>>
>> ########################################################################
>>
>> To unsubscribe from the FSL list, click the following link:
>> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
>
>
> ########################################################################
>
> To unsubscribe from the FSL list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1


########################################################################

To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
<original_bvecs.txt><rotated_bvecs_dsistudio.txt><rotated_bvecs_eddy.txt>


To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1