Grots - maybe the two input images are not well registered or have
very different intensity ranges?
Anyway, this sounds like this is happening when estimating the
inverse data covariance, hmmm.
If the two input images are definitely sensible then maybe this
problem can be gotten around with different initialisation, e.g. by
using the -a option?
Sorry not to be more helpful.....if it might help, we could maybe
make a binary of the new version of FAST (FAST4) available to try.
Cheers, Steve.
On 7 Nov 2006, at 21:36, Rob Roach wrote:
> Steve et al.,
>
> Did as you suggested and installed FSL on 64-FC6. Mfast runs, but
> has new
>
> error (compared to previously crashes on 32 bit OS).
>
> Here is what I used to call mfast, and the results. No images were
> output
> .
>
> Suggestions? Should I upload the images?
>
> TIA,
>
> Rob
>
> mfast -s 2 -c 3 -os -od 03A_T1r_brain 03A_T1r_brain 03A_T2r_brain
> No convergence in 30 SVDCMP iterations
>
> MFAST - Multispectral FAST Version 3.53
>
> Channel 0: /tmp/fsl_hGhMIH_03A_T1r_brain
> Channel 1: /tmp/fsl_hGhMIH_03A_T2r_brain
> Imagesize : 256 x 256 x 154
> Pixelsize : 0.9375 x 0.9375 x 0.9375
>
> initial segmentation by KMeans....
> 4 main iterations ...
> Matrix Inverse: Not invertable!
> Finished
------------------------------------------------------------------------
---
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director, Oxford University FMRIB Centre
FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
------------------------------------------------------------------------
---
|