I want to perform a ROI analysis in the following way:

Create mask with ROI from BOLD dataset 1

Examine activity in BOLD dataset 2

To get the mask, I tried to use the following command:

-3dmerge -prefix block_roi -1tindex 9 -1thresh 10 -1clust_order 5.00 00 decon_1_results+orig

Then, to get a file of the clusters I ran the command:

3dclust -orient LPI -1tindex 9 -1thresh 10 5.00 500 decon_1_results+orig


The problem is the files don't seem to match up. When I look for clusters in the block_roi+orig identified in the file - there are no blobs where the coordinates say they should be and are where they say there aren't.

I'm not sure - but could this also be related to another problem: when I use the 'jump to xyz' command for example, 12, 45, 16, it sends me to another place (15, 27,9)? Have you seen this problem before?


On a related note, I will need to select only one of the areas of activation to be used as the ROI where I want to examine the activity of two different stimulus types. What is the best way to go about doing this? Your ROI notes have left me slightly confused as to the best option for this type of analysis.

Thanks,

Victor