Dear Ella,
I've had a look at your images and I don't think there is a problem. It may be the case that the registration is off by a mm or two at some parts on the cortical surface, but the majority of your activations that appear to be outside of the brain really originate from activations that are also largely outside the brain in the functional space (before any registration). If you look at the example_func and the zstat image in this space you will see that quite a lot of the activations appear at the boundary of the bright signal in the EPI, which is generally surrounding CSF rather than the grey matter, and could be driven from nearby veins. Given you have a resolution of 3mm for your functional images, I don't find that any of the registrations are driving things substantially away in standard space from the locations that you can find in the original functional space. I think this is just what is in your data.
All the best,
Mark
On 28 Aug 2013, at 12:35, Ella Hinton <[log in to unmask]> wrote:
> Apologies. I will upload the original version now. Sorry for the delay.
> thanks again
> Ella
>
>
>
> Hi,
>
> I was hoping that you would upload the original thresh_zstat4 and not the transformed version.
> It is really much more helpful for me if I can see the original activation before transformation and after transformation.
>
> All the best,
> Mark
>
>
> On 21 Aug 2013, at 08:53, Ella Hinton <[log in to unmask]> wrote:
>
>> Dear Mark
>>
>> I think its the latter - in most cases some of the blob is in the brain but the majority isn't in some cases. I'll upload the images for your opinion - thanks again!
>>
>> Ella
>>
>>
>> Dear Ella,
>>
>> Are you talking about a some blobs being located in the grey matter but leaking outside of the brain, or are you talking about ones where the majority of the blob are located outside of the brain? The former case is not a problem and fairly common, but nothing to worry about. The latter case is a more problematic case. If it is the latter case then please send me the thresh_zstat4.nii.gz file corresponding to the registration files you sent before and I can have a look at them.
>>
>> All the best,
>> Mark
>>
>>
>> On 20 Aug 2013, at 11:51, Ella Hinton <[log in to unmask]> wrote:
>>
>>> Hi Mark,
>>>
>>> Thanks for your email. Its good to know that you think the initial registration is good.
>>>
>>> With regards to viewing the thresholded zstat images - I have tried using applywarp just as you suggested. Unfortunately I am still getting activation blobs outside of the standard brain on all the images. After reading the applywarp information on the website, I tried adding a premat file (example_func2highres.mat in this case), but that made it much worse! I used the verbose option in case any errors came up: this is the command and output:
>>>
>>> [psech@adara mando_opt2A.feat]$ applywarp --in=thresh_zstat4.nii.gz --ref=reg/standard.nii.gz --warp=reg/example_func2standard_warp.nii.gz --out=thresh_zstat4_standard_warp.nii.gz --verbose
>>> Automatically determining absolute/relative warp convention
>>> Assuming warps was relative
>>> superflag = 0
>>> ssvec = 0 0 0
>>>
>>> I would be really grateful for any further ideas for me to try to register these images!
>>> With thanks
>>> Ella
>>>
>>> ********
>>>
>>> Hi,
>>>
>>> The registrations look good to me, and I can definitely see the difference in applying the fieldmap, so I'm surprised to say that you thought that changing the sign didn't make a difference. I would double-check that.
>>>
>>> I didn't spot an obvious issue in your previous email (and please keep copying them in the thread as it helps in answering them). You previously said that you used the command:
>>> flirt -in thresh_zstat1.nii.gz -ref reg/standard.nii.gz -applyxfm -init reg/example_func2standard.mat -interp trilinear -out thresh_zstat1_standard.nii.gz
>>>
>>> but this is only good if you have just done linear registration. If you have done nonlinear registration then you need to use applywarp with the example_func2standard_warp.nii.gz image with a command like:
>>> applywarp --ref=reg/standard --in= thresh_zstat1 --warp=reg/example_func2standard_warp --out=thresh_zstat1_standard
>>>
>>> All the best,
>>> Mark
>>>
>>>
>>> On 15 Aug 2013, at 22:38, Ella Hinton <[log in to unmask]> wrote:
>>>
>>>> Hi Mark
>>>>
>>>> Many thanks for your email. I have uploaded the reg folder and design.fsf file as requested. Let me know if the upload hasn't worked ([log in to unmask]). I really appreciate you taking a look.
>>>>
>>>> The uploaded data are from the Feat we ran with the default unwarp direction of -y. We also ran it with unwarp direction of y, but it didn't appear to make any difference.
>>>>
>>>> Thanks in advance
>>>> Ella
>>>>
>>>> (Research Associate, CRICBristol)
|