[Neuroimaging] Understanding atlas usage

[bthirion]

On 30/11/2018 17:26, Alessandro Stranieri wrote:
> Hi,
>
> I am working on a small project with the adhd data. My background is 
> software engineering, but I am very new to nipy and neuroscience in 
> general so I might get some terminology wrong.
>
> Currently I am working my way through the correlation/connectome 
> examples. I have a few doubts but, in order to keep it simple, I will 
> just
> post a couple, hoping for help.
You're probably using Nilearn. Could you post these questions on 
Neurostars.org, so that other people can benefit from the answers ?
>
> 1. The fetch_adhd functions states that a maximum of 40 subjects can 
> be retrieved. This means that if I want to use more, I need to do all 
> the processing myself from the files at
> https://www.nitrc.org. Is it correct?
>
Yes.
> 2. nilearn provides some atlases and I have more or less understood 
> how to use them. However, on NITRC one can also find 2 functional 
> parcellation templates, and time series are provided
> for those parcellations. I think I managed to create and display 
> connectomes with those templates but, what if I want to know the label 
> of a region? Those regions are labelled by numbers.
> For example, if I use the aal atlas, I can see 'Precentral_L', 
> 'Precental_R' and so on. Can I do the same with the cc200 for example? 
> I fear that, since those parcellation are functional
> and not anatomical, there are no real labels associated to the region. 
> But I would still like to have a data-structure that allows me to 
> query things like: this region is in V1, or in the
> frontal lobe or other.
>
The best way to learn about parcellations is to run the corresponding 
examples of the library.

Each atlas comes with different information, so you may not have 
functional labels if the atlas comes from a functional parcellation, 
such as resting-state for instance.

HTH.

Bertrand

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20181130/06664d5b/attachment.html>