Track to region questions

[corin...@gmail.com]

Hi Frank,

Prior to beginning the track to region connectome analysis, I have two questions that I am hoping you can answer.

1. Can you please confirm whether it is possible to import a set of atlas-derived ROIs already in subject space? I.e., the subject specific freesurfer (or other) parcels?

2. For the percentage of voxels within each label/region through which the tract passes (or ends), is there a way to control for the volume (or # voxels) of that label/region? Or will this need to be done after the fact in an external statistics package? Wondering as this may be an issue when investigating the outcomes between patient and control groups where regional volume changes are anticipated to occur.

Best

Corinna

[Frank Yeh]

Hi Corinna,

Yes, you can use subject-specific atlas ROIs, including FreeSurfer
parcels, as long as they are already aligned to the diffusion/subject
space. You can load them as regions rather than using an MNI atlas.

For your second question, DSI Studio's tract-to-region "intersect
ratio" already normalizes the number of intersected voxels by the
number of voxels in the region. I checked the current implementation
to confirm this. So this metric is already adjusted for ROI size.

That said, if regional atrophy itself is an important group effect, I
would still keep the regional volume as a separate variable/covariate
in the statistical analysis rather than relying only on the normalized
intersect ratio. This lets you distinguish altered tract-region
involvement from the volume change itself.

Best,
Frank

> --
>
> You received this message because you are subscribed to the Google Groups "DSI Studio" group.
>
> To unsubscribe from this group and stop receiving emails from it, send an email to dsi-studio+...@googlegroups.com.
>
> To view this discussion visit [https://groups.google.com/d/msgid/dsi-studio/e207b206-8687-43e9-b256-70399b0538ban%40googlegroups.com](https://groups.google.com/d/msgid/dsi-studio/e207b206-8687-43e9-b256-70399b0538ban%40googlegroups.com?utm_medium=email&utm_source=footer).