Dear Kathleen,
On 13 Nov 2019, at 17:02, Hupfeld,Kathleen E wrote:
> Dear Dr. Gaser,
>
> Thanks so much for your response. I had previously tried the geodesic
> shooting registration as well, which yielded very similar results to
> default dartel warping. I agree that the warping results I shared are
> generally very reasonable (& we don't have any specific hypotheses
> linked to basal ganglia). However- one other question, with regard to
> ANTs: would it be problematic to run an additional warping step on the
> modulated GM segments (mwp1*.nii)?
although the additional registration by ants is used to better register
the image, the applied deformations are not used to also change the
modulation of the image. Therefore, I would not recommend this
additional step, because you locally change the volumes by applying the
deformations.
Best,
Christian
>
> I have done the following steps:
> 1) Run the CAT12 T1 template (Template_T1_IXI555_MNI152_GS) thru CAT12
> to get a template mwp1*.nii file
> 2) Use ANTs SyN to warp each subject's mwp1*.nii segment to the CAT12
> T1 template mwp1*.nii segment
>
> This has worked very well in that the basal ganglia line up very
> nicely and the ventricle size is very homogeneous across young and
> older adults. Are there any problems with adopting this method of
> using ANTs after CAT12?
> [cid:21c3b21f-5c49-4d4f-8610-6277806d3c11]
>
> Any advice would be greatly appreciated.
>
> Best,
> Kathleen
> ________________________________
> From: Christian Gaser <[log in to unmask]>
> Sent: Monday, November 11, 2019 5:54 PM
> To: [log in to unmask] <[log in to unmask]>; Hupfeld,Kathleen E
> <[log in to unmask]>
> Cc: Christian Gaser <[log in to unmask]>
> Subject: Re: CAT12 warping in aging subjects
>
> [External Email]
>
> Dear Kathleen,
>
> On Thu, 31 Oct 2019 17:18:48 +0000, Hupfeld,Kathleen E
> <[log in to unmask]> wrote:
>
>> Dear SPM experts,
>>
>> I'd like to use CAT12 to compare gray matter volumes and run
>> brain/behavior correlations in a sample that includes ~30 young and
>> ~30 older adults. I wanted to know if anyone had thoughts regarding
>> the DARTEL warping used in CAT12 with aging subjects. Comparing the
>> modulated, normalized T1s (wm*.nii) and the gray matter segments
>> (mwp1*.nii) between subjects, the ventricles are still slightly
>> larger in all of the OA subjects.
>>
>> E.g., below are screenshots of a young (left) and older (right)
>> participant. Ventricles are larger on OA, so the crosshair is on the
>> edge of gray matter for the young adult, but inside the ventricle for
>> the old adult.
>> Will this be generally problematic for my study?
>
> Even of there remain some differences in the ventricles after spatial
> registration, the GM overlap looks fine except for the caudate
> nucleus. This might be only an issue if your hypothesis is linked to
> that region. However, you can also try the geodesic shooting
> registration (probably with 1mm voxel resolution) to see whether the
> overlap in these regions will improve.
>
>>
>> We have used ANTs
>> (https://urldefense.proofpoint.com/v2/url?u=http-3A__stnava.github.io_ANTs_&d=DwIFaQ&c=sJ6xIWYx-zLMB3EPkvcnVg&r=wzk53wbbZ2Sk91GnOt-W5Q&m=kXOzUm0BfYgThodhnkMweerIkeHEAFNEzhflxnZnMN8&s=zIUBYr0H_7xsDOAZ4zQYLo4N0Bp19iS2a9CjR2lXlHM&e=
>> ) previously for normalizing T1s and fMRI images to MNI space in
>> aging populations, which has performed very well for handling issues
>> like large ventricles for old adults; however, given the large number
>> of steps that run during CAT12's "segment" procedure, I'm not sure
>> that ANTs could be interfaced well with CAT12?
> It will be probably too complicated to use the ANT registration for
> CAT12...
>
> Best,
>
> Christian
>>
>> Any thoughts would be greatly appreciated.
>>
>> Best,
>> Kathleen
>>
>> Modulated, normalized T1s:
>> [cid:4cf6ab88-4dd3-4db5-aa40-eaefd693de8e]
>>
>> Gray matter segments for same two people, same coordinate:
>> [cid:6e5e13e7-abd2-4825-931c-33b9f166c147]
>>
>>
>>
>> Kathleen Hupfeld
>>
>> PhD Candidate, Biobehavioral Science
>> Neuromotor Behavior Lab
>> Department of Applied Physiology & Kinesiology
>> University of Florida
>>
|