Wow, that's interesting. In both cases it's called with the default settings, without specifying any special options on the command line. You could verify this by running the respective versions of bedpostx independently of tracula and looking at the outputs. Unless of course something has changed with the inputs to bedpostx (so in the pre-processing steps), in which case the differences wouldn't have to do with bedpostx itself.
On Wed, 12 Feb 2014, jwaugh@nmr.mgh.harvard.edu wrote:
Thank you Anastasia. Unfortunately, I posted the question to the FSL list first and after a few go-rounds with clarification and more info, they felt like it was potentially a problem of the calls to the bedposts command from within Freesurfer. Their final assessment was that there should be no differences in output between different versions of bedpostx, so there must be some difference in how stable 5.1 and 5.3 call the command. Can you provide any more details about how 5.1 and 5.3 differ in this way?
Thank you, Jeff
Hi Jeff - Questions about bedpostx output files are best addressed to the FSL list, as it's an FSL program that is not developped by us. The appropriate info to provide them is the FSL (rather than freesurfer) version numbers.
a.y
On Tue, 11 Feb 2014, jwaugh@nmr.mgh.harvard.edu wrote:
Hello all,
I have run bedpostx on a series of DTI scans over ~1 year, the earlier ones in stable 5.1 and later ones in 5.3. I'm seeing quite a lot of variability in intensity from the earlier to later versions - re-running bedpostx in 5.3 on an older scan and comparing with 5.1 yields up to 10% variation in intensity at any given voxel between versions. I've been comparing the mean_f1 images by subtracting the (new - old) using fslmaths. The differences are not isolaed to surfaces, but can also be deep in the white matter or parenchyma. Has anyone else seen this?
Thanks, Jeff _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer