Hi, Alaka,

I used to do similar VBM analysis using SPM. I suggest you to try SPM related toolboxs called "Marsbar" or "WFU_Pickatlas" to perform ROI analysis and extract desired ROI volume individually. Hope it will help, good luck!

Tianjiao

Dear Alaka,

It is not perfectly clear if you are referring to 1) perform analysis inside a ROI or to 2) extract results inside a ROI. However, in AFNI:

1) if you want to perform analyses only within a ROI / a predefined volume you can include "-mask MASKNAME" among the analysis options (e.g. 3dLME). Remember that the number of voxels analysed may impact the subsequent multiple comparison / cluster correction, so use it wisely;

2) more probably, you want to extract values from a ROI:

"3dROIstats -nzmean -nomeanout -quiet -mask MASKNAME FILENAME > OUTPUT.1D"

where "-nzmean" and "-nomeanout" ensure zero-valued voxels to be ignored, MASKNAME is the name of you ROI file, FILENAME is the name of your input file (e.g. task-related activity betas), and OUTPUT.1D is the output text file.

You may also want to look at the program 3dTcorr1D, which may be useful for your purposes (correlation) given that sometimes using predefined ROIs may bias the analysis (https://afni.nimh.nih.gov/pub/dist/doc/program_help/3dTcorr1D.html).

Hope this helps!

Simone

Tianjiao Li Thank you so much. I have never tried those. I will certainly give it a try

Simone Di Plinio Thank you so much for your help. AFNI is more familiar to me, so I will try it.

CAT12 toolbox of SPM12 may facilitate your analysis. If you know the routine SPM workflow, CAT12 is very straightforward.

http://www.neuro.uni-jena.de/cat/

Dear Alaka Acharya,

If you use CAT12 in expert mode, you will get options to extract regional volume using different atlas. Have a look on this manual http://www.neuro.uni-jena.de/cat12/

If you want to extract volume for a particular coordinate then you need to create a sphere mask using WFU-Pickatlas and then extract volume using get_totals([],[],[]) in MATLAB.

Best