Anatomical volumes constructed from only eight slices are not really ideal for
working with in SPM (although if you actually have 82 slices, then the data
may be OK to work with).
I can not comment on the behaviour of BET. However, if the data are
skull-stripped before segmenting, and you make sure that the skull-stripped
images are oriented correctly (use Check Reg with your skull-stripped images
and the template data supplied with SPM), then things should proceed more
easily.
Best regards,
-John
On Monday 24 November 2008 15:08, 李猛 wrote:
> Thank you for your prompt reply! But I am still a little puzzeling on this
> issue.
>
> By using MRIcroN's dcm2nii tool, we convert original GE dicom files to
> spm5(3D NIFTI hdr/img) format. The dimensions are 512*512*8 at this
> time.With fractional intensity threshold value of 0.15, we use bet tool
> within MRIroN to strip skull.Output file dimensions become 512*512*82.Maybe
> this lead to the initial import problem. Because I find the right bet
> results usually look like from 512*512*152 to 155*270*260.The X and Y
> dimensions should decrease.However,I don't know how to fix it?Or is there
> other bet tool?
>
> Thanks.
>
>
> 2008/11/24 John Ashburner <[log in to unmask]>
>
> > If the orientation of the imported images is totally wrong, then I would
> > guess
> > that the tissue probability maps were not well aligned with the images
> > that were being segmented, and that the c1, c2 images were not as
> > accurate as they
> > could be.
> >
> > Best regards,
> > -John
> >
> > On Saturday 22 November 2008 06:26, 李猛 wrote:
> > > Hi,
> > >
> > > We are using DARTEL to create template.
> > >
> > > First, we reorient all T1 images and get the acceptable c1,c2 images.
> >
> > After
> >
> > > finishing initial import, we get the weired rc1,rc2 images.
> > >
> > > It looks like the orientation have a big offset and the axis changed
> > > from xyz to xzy.
> > >
> > > Now, What can be done to avoid or reduce this effect?
> > >
> > > Thanks
> > >
> > > --
> > > albert
> > > Lab of Complex System and Artificial Intelligence,
> > > Institute of Automation,Chinese Academy of Sciences(CASIA)
> > > Beijing, 100190,P.R.CHINA
>
> -- albert
> Lab of Complex System and Artificial Intelligence,
> Institute of Automation,Chinese Academy of Sciences(CASIA)
> Beijing, 100190,P.R.CHINA
|