Functional MRI real anatomy

Discussions
,
1

Hi everyone!
Is the first time I'm trying to use subjects MRI instead protocol's default anatomy template,
I'm importing DICOM images with spm12,

But I have a more theorethical question:
For this subject I don't have the T1 sequence, I just have T2 and some fMRI from emotional task,
is it ok to use one of that sequences to import the anatomy? will be any differences if I used the T1 instead fMRI sequences (I mean, I just need the anatomy)?

Now I'm importing some fMRI sequence but I dont know if that is ok

Thanks in advance!

2

Hi Tourette,

The fMRI sequence usually has a way lower resolution than a T1 or a T2 sequence. Most segmentation programs work directly on the T1 and optionally use the T2 sequence to improve skull segmentation and some other sequence specific details.

If it works in your pipeline, you might want to check the resolution of your anatomy and the segmentation quality.

Kind regards,
Steven

3

I'm not sure whether FreeSurfer/BrainSuite/BrainVISA/CAT12 can generate cortical surfaces based only on T2 images. I have never seen it...
@CGaser @Anand_Joshi @tmedani ?

I'm not sure you'd be able to use something else than an MNI template if you only have a T2 image of your subjects... Alternatives can be:

https://neuroimage.usc.edu/brainstorm/Tutorials/LabelFreeSurfer
https://neuroimage.usc.edu/brainstorm/Tutorials/SegBrainSuite
https://neuroimage.usc.edu/brainstorm/Tutorials/SegBrainVisa
https://neuroimage.usc.edu/brainstorm/Tutorials/SegCAT12

4

Although CAT12 can handle T2 weighted images the resulting quality is usually quite inferior compared to T1 weighted data. Often T2 images has lower spatial resolution, but the main issue is that the contrasts between grey and white matter are rather low which is mandatory to obtain a good segmentation.

Best,

Christian

5

Thanks for the reply!
Finally I used a strange MRI sequence "AX 3D SPGR", I'm not sure what it is but it seems to work,
with another patient I'm gonna try with FLAIR,
I'm performing analysis with these surfaces and letting you know if that works
Thanks!