Hi Shruti,

As there is one *_Z.nii.gz file per subject (with all ICs for that subject), you may find it easier to use fslstats. The option -M gives the output across the whole brain, the -k allows supplying a mask, and -t allows one measurement per timepoint (i.e., per IC) from the *_Z.nii.gz files. If you use a mask that is specific for a given IC, consider disassembling these _Z files (with fslsplit), then running fslstats on each separately (or merging across subjects with fslmerge, then using fslmeants).

You'll certainly need to script this, e.g., with bash.

All the best,

Anderson

On 25 April 2016 at 10:23, Shruti Narasimham <[log in to unmask]> wrote:

Thank you for all your help Anderson!

I would like to obtain z-values / quantify the results produced by randomise.

More specifically, I'd like to

Extract Mean z-values for each subject from the individual default-mode network (DMN) z-maps obtained from the dual regression analysis, compare using independent-samples t tests and plot these in the form of boxplots/bar graphs.

Thus, for these I am trying to find out how to extract these z-values for each IC / RSN / subject from the z-maps of the DR analysis.

Any help regarding this will be greatly appreciated.

Thank you!

Regards,
Shruti

tests

On 24 April 2016 at 12:07, Anderson M. Winkler <[log in to unmask]> wrote:
Hi Shruti,

For (1): These are the results produced by randomise.

For (2) and (3): I don't quite follow what you would like to do. Could give details?

All the best,

Anderson

On 23 April 2016 at 16:16, Shruti Narasimham <[log in to unmask]> wrote:
Hello,

I have run my dual regression and randomise after performing group ICA on a set of data consisting of 3 groups - patients, controls and relatives. I have identified and grouped ICs according to specific RSNs.

I now want to extract mean z-values from each individual spatial map and compare RSNs FC between these groups.

How do I -

1. Identify regions that show significant differences in FC between groups?

2. Extract mean z-value per subject for each RSN?

3. Find additional mean z-zcores of significant voxels per subject for networks that show differences in FC?

Please do let me know how to perform this analysis.

Regards,
Shruti

--
Shruti Narasimham
PhD Candidate
IRC Scholar 2015-2019
Trinity Centre for Bioengineering
Department of Electronic & Electrical Engineering
Trinity College Dublin
Dublin 2- Ireland