---------------------------------------------------------------------------------------Original
Message-----
From: SPM (Statistical Parametric Mapping) [
mailto:[log in to unmask]] On Behalf Of
Chaitee Sengupta
Sent: Tuesday, May 03, 2005 4:27 PM
To:
[log in to unmask]Subject: [SPM] small volume ROI analysis and search
volume
We are doing an fMRI ROI analysis using manually created (in
MRIcro) ROI image masks and the WFU pickatlas interface with SPM2 utilizing SVC.
We are finding that the volume our ROIs as drawn and the search volume reported
on the SPM results do not match.
Here are the steps we have
followed:
1. Draw ROI on normalized structural image (2x2x2mm voxels) in
MRIcro 2. Export ROI as Analyze image 3. In SPM Results, use ROI analysis (WFU
Pickatlas interface), specify mask image file.
In one case the ROI we
drew had 119 voxels, but the SPM2 analysis search volume was reported as 113
voxels with resels reported as 0.3 . In another case the ROI was 86
voxels, but the SPM2 search volume was 23 voxels with 0.1 resels.
Why are
we seeing the difference between the original ROI and the search volume? How
does SPM calculate the search volume when doing an ROI
analysis?
Thanks,
Chaitee Sengupta
Research
Assistant
Cognitive Neuropsychology Lab
University of Washington,
Seattle