The atlas images from Shen et al Neuroimage 2013 are in MNI space, correct?

Originally posted by Maxwell Bertolero: Also, what's the difference between the two sets of parcellations? 

The two versions are one at 1mm and one at 2mm resolution. They are in MNI space.

Thanks! Sorry, to be clear I meant the one used in the nature neuroscience paper and then the ones from the neuroimage paper. Are they both in MNI space?

The neuroimage atlas was generated using older low res data. The natural neuro one used multi-band data - this is the Oct 2015 version.

Okay great, but they are both in MNI space?

Thanks!

Yes

I am using this atlas to analyze subjects that were normalized to the MNI152 template included with FSL. Is this atlas in the correct space for this, even though it was made in SPM?

When I pull up the atlas with this template in fslview, it aligns quite well, I just wanted to make sure. 

Thanks!

Simply put, this is MNI152, right?

I need to confirm. It maybe MNI 27 which is simply but not identical.

Thank you. Would it be reasonable to use with MNI152 even if it was made for MNI27?

I am also referring to the neuroimage parcels.

Probably. I am out of the office, let me get back to you Monday once I confirm.

PS: All of these parcels are always going to be sometimes a better and sometimes a worse approximation. Registration is never perfect either but at the resolution of the parcel size it should be close enough.

So you're saying that The Colin27 and MNI152 are similar enough regarding the x y z coverage so, at the level or parcels, any small differences are glossed over, especially after registration errors? That makes sense.

Okay, when I overlay the atlas, it does look like it was fit on the Colin27, but the differences between how it fits Colin27 and MNI152 are very minor, so I think it's fine for use with subjects registered to MNI152, but it would be nice to get someone else's opinion on this matter.

I think the easiest way to check how much difference it will make is to register the Colin 27 to MNI 152 and propagate the template over - (use a very smooth nonlinear registration probably) and to see how much difference does using this resliced atlas (remember to turn interpolation off when reslicing the atlas) makes in the final results in a subset of your subjects.

Hello,

I wonder whether is possible to select just some ROIs from the Shen Atlas in order to observe the FC among specific brain areas, such as DMN, salience, etc. Of course I could make my owm ROIs based on the MNI coordinates  I get after getting the ROI book. But the parallelization (e.g. ROI size) would be completely arbitrary. So it would not make sense for me. Is there any way to do such "networks selection" taking advantage the parallelization proposed by the atlas?

Thanks in advance