Print

Print
I don't have a solution for SPM8 - other than using something like MRIcron to manually edit the results.  If voxels have the same intensities as GM, and are roughly where you would expect to find GM, then they will be classified as GM.

Besides, the results from New Segment in SPM8 are not as accurate as they could be - particularly for more atrophied brains.  This aspect was fixed for SPM12.

Best regards,
-John



On 2 June 2014 17:55, Andrea Gold <[log in to unmask]> wrote:
Hello,

I am running a VBM analysis in SPM8, following the tutorial (link: http://www.fil.ion.ucl.ac.uk/~john/misc/VBMclass10.pdf) and using New Segment, Run DARTEL (Create Template), and DARTEL tools-> Normalize to MNI space.

There is some remaining dura around the grey matter segmented images after running 'New Segment' and I am writing to ask whether there are any steps within SPM8 New Segment to adjust to address this? The message copied below did not have a response, and other posts in the listserve archives said to try the VBM8 toolbox or SPM12b. Does this mean there are no other options for removing leftover dura/non-brain within SPM8 New Segment? (other than changing the Bias Regularisation from the default of 'very light' to 'extremely light', which did not fix the issue in my case).

Thanks,
Andrea

--

I'm running into problems with the New Segment procedure in SPM8.  I keep on obtaining a fine ring of dura around the grey matter native segment image even after messing with the bias and regularization parameters.  The data were collected on a 3T and are good quality.  I don't run into the problem with the old segment.  Any way to get around this problem?   I'd like to use the New Segment to DARTEL routine.

Thanks to any and all,
Jeff Browndyke