Hi,
Yes, boundary correction does affect the volume size. You can specify the
boundary value
correction for first_utils using "-p 3". It should really only be used
when reconstruction in native space with no alignment, otherwise the
intensity sample will be from the wrong location in the image space.
The flirt matrices should be input as a text file list pointing to the
flirt matrices
Cheers,
Brian
> I looked at the source, and it definitely does affect the volume size.
>
> Which leads me to another issue: I haven't actually gotten first_utils to
> work on anything other then MNI space. I repeatedly tried
> --useReconNative, and tried to feed in flirt matrices ... However I
> continuously receive errors.
>
> Either:
> Unrecognised binary matrix file format - St9bad_alloc
> 'NEWMAT::SingularException'
>
>
> Here was my function call (does something look obviously wrong?):
> first_utils --usebvars --vertexAnalysis -i tmp_R_Amgy.bvars -d design.mat
> -o
> test --useReconNative -f
> /data/mireccAnat/data/20070122_32684/firstSeg/T1_flip_to_std.mat
> /data/mireccAnat/data/20070417_32965/firstSeg/T1_flip_to_std.mat
> --useRigidAlign
>
> This was just a test to see if I could run 2 subjects through ... Scaled,
> with MNI space, and rigid align have always worked for me.
>
> -----Original Message-----
> From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On Behalf
> Of Peter Kochunov
> Sent: Friday, January 25, 2008 11:31 AM
> To: [log in to unmask]
> Subject: Re: [FSL] FIRST and subcortical structures volumes
>
> It certainly does, at least in my experience.
> pk
> ----- Original Message -----
> From: "Christopher Petty" <[log in to unmask]>
> To: <[log in to unmask]>
> Sent: Friday, January 25, 2008 10:27 AM
> Subject: Re: [FSL] FIRST and subcortical structures volumes
>
>
> how does the boundary correction value affect the output volume size?
>
> i've noticed that my volumes from fslstats -V are quite different from
> those
> in the .vols output of first_utils.
>
> ie: L-amyg = 2626 voxels in .vols and 2019 voxels from fslstats
>
> also, does the surface reconstruction method and scaling from the vertex
> analysis also affect the volume sizes in the .vols output?
>
> thanks,
> -chris
>
>
|