The recent paper by Wonderlick et al. (Neuroimage 44:1324-1333, 2009) is relevant to your situation. They show that differing voxel geometry can definitely lead to bias in the results (as Bruce noted).
cheers, Mike H.
On Tue, 2009-03-17 at 21:26 -0600, Jeff Sadino wrote:
Hi Bruce,
Yes, we are going to combine them into a single study. Is there a place on the wiki I can read about how to error check for these problems?
Thank you, Jeff
Date: Tue, 17 Mar 2009 19:39:34 -0400 From: fischl@nmr.mgh.harvard.edu To: jsadino@hotmail.com CC: freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] Changing from 1.4x1.4x1.4 mm slices to
1x1x1
Hi Jeff,
are you intending to combine them into a single study? This can be problematic as you'll get differential sensitivity and possibly
bias.
cheers, Bruce
On Tue, 17 Mar 2009, Jeff Sadino wrote:
Hello,
I am analyzing some subjects from several years ago when we had a
protocol that called for slices 1.4x1.4x1.4mm. I am also including in the study more recent scans with a protocol for 1x1x1mm slices. I am using freesurfer 3.0.5. I am wondering if the different slice dimensions will cause differences in the analysis. As I read on a previous post, freesurfer resamples all its input data into 1mm isovolumetric voxels anyways, so I should be ok?
Regards, Jeff Sadino
Hotmail® is up to 70% faster. Now good news travels really fast. http://windowslive.com/online/hotmail?
ocid=TXT_TAGLM_WL_HM_70faster_032009
Windows Live™: Keep your life in sync. Check it out. _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer