Thank You very much!
Monica G Chica
PhD student Biomedicine
Sherman Health Science Research Centre (Canada)
University of Barcelona (Spain)
www.monicagiraldochica.com
On 2012-10-22, at 7:03 AM, Mark Jenkinson <[log in to unmask]> wrote:
> Hi Monica,
>
> The orientation is correct if the labels in FSLView are in the right place (L on the left, P on the posterior end, etc.)
>
> If the labels are fine for the LGN mask, then you can apply fslreorient2std directly to the LGN mask too.
>
> All the best,
> Mark
>
>
>
> On 22 Oct 2012, at 11:58, "Monica G. Chica" <[log in to unmask]> wrote:
>
>> Hi, thank you very much for your response. I just want the brain to be aligned the same way as the ones in standard space, all straight and no angle. So I guess fslreorient2std is the one. I just have two questions regarding that:
>>
>> - Saad said I can use it if the orientation of my header is ok. How do I check that? (sorry if is an obvious question).
>>
>> - after I use fslreorient2std with my PD image, can I apply the same "movement" to my LGN masks? Because I want the mask to be positioned correctly as it's corresponding structure in the proton density.
>>
>> Thanks,
>>
>> Monica G Chica
>> PhD student Biomedicine
>> Sherman Health Science Research Centre (Canada)
>> University of Barcelona (Spain)
>> www.monicagiraldochica.com
>>
>> On 2012-10-22, at 4:52 AM, Mark Jenkinson <[log in to unmask]> wrote:
>>
>>> Dear Monica,
>>>
>>> There is no such thing as staying in native space *and* having standard space *coordinates*. If what you want is to have the axes aligned in the same way as the ones in standard space (x in left/right, y is A/P, z is I/S) then you can use fslreorient2std. Don't use fslswapdim or fslorient, as fslreorient2std is specifically designed to do exactly this function and is a lot simpler and less likely to cause any problems of flipping your data incorrectly than if you used fslswapdim or fslorient.
>>>
>>> The information in the sform/qform (from the dicom) can also be used by using the flag -usesqform in FLIRT along with -applyxfm in order to resample your image along the *scanner* axes, which can be better if your acquisition is oblique (since fslreorient2std will only ever do 90/180/270 degree rotations).
>>>
>>> If you really want to be able to use the *coordinate* information then you will need to use flirt and fnirt, but this will not keep you in native space.
>>>
>>> If you want to use a rigid body transformation (6 dof, with rotations and translations only) to align the anterior commissure, the ac-pc line, and the inter-hemispheric plane, then you can use the script aff2rigid, which takes a 12 dof transformation from FLIRT (the typical standard space transformation) and calculates the best 6 dof fit that satisfies the above set of constraints. You can use this to then transform your image "into" standard space, but preserve the original dimensions of your native brain.
>>>
>>> So the answer depends a lot on what you want, and this was not clear from you question. If this hasn't answered your question, or you need more details, then please email back with more information.
>>>
>>> All the best,
>>> Mark
>>>
>>>
>>> On 22 Oct 2012, at 09:18, Saad Jbabdi <[log in to unmask]> wrote:
>>>
>>>> Hi - if your orientation info in your header are correct, you can use fslreorient2std.
>>>>
>>>> Cheers
>>>> Saad
>>>>
>>>>
>>>>
>>>> On 22 Oct 2012, at 03:32, Monica Giraldo wrote:
>>>>
>>>>> Hello,
>>>>>
>>>>> Is there a tool in FSL to orient the brain into standard space coordinates? Without normalizing the brain, keeping it in native space.
>>>>>
>>>>> Thank you very much,
>>>>>
>>>>
>>>> --
>>>> Saad Jbabdi
>>>> University of Oxford, FMRIB Centre
>>>>
>>>> JR Hospital, Headington, OX3 9DU, UK
>>>> (+44)1865-222466 (fax 717)
>>>> www.fmrib.ox.ac.uk/~saad
>>>>
>>
|