[Neurodebian-users] FSL FNIRT bug

[Michael Hanke]

Hi,

I'm looking into this, please see inline comments below:

On Thu, Aug 25, 2011 at 08:32:29PM -0400, Jan Scholz wrote:
> Hi,
> 
> I seem to have found a bug in FNIRT, FSL's non-linear alignment tool.
> For some reason it does not save the affine transformation if a
> non-linear initial warp (inwarp) is used to initalize FNIRT. You can
> test this easily by running the following commands (<5min). The error
> does not come up on a CentOS system with the newest build,
> non-neurodebian). So it might have been fixed in some newer version
> (FNIRT build 419).

Hmm, strange. I contacted upstream to figure out where that build is
coming from -- maybe we need to update the package. However, the latest
version comes with 418....

> I checked the same commands on neurodebian ubuntu lucid and
> latest-version locally compiled CentOS, and they produce different
> results, if I'm not mistaken. Please check whether you can reproduce
> the bug with the following commands.

Thanks a lot for the commands. Would you be willing to turn this into a
(shell) script that runs these commands and fails whenever there is a
bug and passed whenever it works. If possible, using an input image that
already comes with the FSL sources. Since you have both, a working and a
broken version, you are in the ideal position to do that.

The reason why I'm asking is this: We are collecting such little
test-snippets and want to turn them into a richer test suite for FSL.
Such suite would allow us to better check correct performance of
upcoming releases. If you send us that script, we'll make sure that this
bug never hits you again! ;-)

Thanks in advance,

Michael

-- 
Michael Hanke
http://mih.voxindeserto.de