Dear Heidi,
I've received your structural and transformed example_func, but I really need to have the original example_func as well.
Additionally, it would be very helpful if you supplied the registration files for this subject.
If you could zip up the first level reg directory (form within the feat directory) and upload this, then I should be able to get a better feeling for what is happening.
All the best,
Mark
On 30 Apr 2014, at 10:38, Mark Jenkinson <[log in to unmask]> wrote:
> Hi,
>
> This sounds unusual.
> Can you send us data from an example subject so that we can see the problem ourselves?
> You can send the data to:
> https://oxfile.ox.ac.uk/oxfile/work/extBox?id=68312615463381F4C
>
> All the best,
> Mark
>
>
> On 28 Apr 2014, at 14:16, [log in to unmask] <[log in to unmask]> wrote:
>
>> Hello Mark,
>>
>> Thank you for the reply and for your insight.
>>
>> I have checked the highres2standard_linear images and compared these with
>> the oriented highres and the non-linear images. I also doubled checked my
>> naming convention/directory structure and confirmed that that was not a
>> contributing factor.
>>
>> It seems that the registrations are distorting/dropping out the wide sulci
>> areas during the non-linear transformation. Any ideas on why this may be
>> happening and how I may be able to fix these registrations?
>>
>> Please let me know if you would like me to send you additional screenshots
>> of these different registration step comparisons.
>>
>> Thank you again for the help.
>>
>> Jacob
>>
>>
>>
>>> Hi,
>>>
>>> The first thing to check is if the problem arises from the linear or the
>>> non-linear registration steps.
>>> Have a look at the highres2standard_linear images and see if these are OK
>>> or not.
>>>
>>> If these are fine then the problem is in the non-linear stage (FNIRT) and
>>> may be due to it being unable to find the non-brain extracted image (as
>>> both images should be in the same directory and follow the naming
>>> convention: aname.nii.gz and aname_brain.nii.gz, where the "aname" part
>>> can be anything you want).
>>>
>>> If the linear step is problematic then it might be due to bias field
>>> (which can be reduced/removed with FAST), or large initial orientation
>>> difference (which can be solved by running fslreorient2std on all images
>>> first).
>>>
>>> There are other possibilities too, but these are the things I would look
>>> at first.
>>>
>>> All the best,
>>> Mark
>>>
>>>
>>>
>>>
>>> On 25 Apr 2014, at 13:53, [log in to unmask]
>>> <[log in to unmask]> wrote:
>>>
>>>> Hello FSL Experts,
>>>>
>>>> In reviewing structural and functional registrations from within
>>>> subjects'
>>>> higher-level FEAT directories, I have come across two subjects where
>>>> FLIRT/FNIRT seems to have made an error in determining the brain
>>>> outline.
>>>> I make this assumption based on the way the images seem to be distorted
>>>> relative to native space folding patterns that contain a large amount of
>>>> space.
>>>>
>>>> I have attached screenshots of the two subjects' structural images in
>>>> native and MNI space and am interested to hear what you may think is
>>>> going
>>>> on, and of course, how best to fix the registration.
>>>>
>>>> Please let me know if you would like me to send additional information
>>>> or
>>>> images and I would be happy to do so.
>>>>
>>>> Thank you for the help and insight.
>>>>
>>>> Jacob
>>>>
>>>>
>>>>
>>>>
>>>> The information in this e-mail is intended only for the person to whom
>>>> it is
>>>> addressed. If you believe this e-mail was sent to you in error and the
>>>> e-mail
>>>> contains patient information, please contact the Partners Compliance
>>>> HelpLine at
>>>> http://www.partners.org/complianceline . If the e-mail was sent to you
>>>> in error
>>>> but does not contain patient information, please contact the sender and
>>>> properly
>>>> dispose of the e-mail.
>>>> <Subject1.jpeg><Subject2.jpeg>
>>>
>>>
>>>
|