You can obtain a copy of the latest version of the MNI_AutoReg software using anonymous FTP from:
x.yyis the release number).
MNI_AutoReg requires that both the MINC and netCDF packages be built and installed first. The MINC file format is a highly flexible medical image file format built on top of the netCDF generalized data format. You should be able to get them both from
You will need MINC version 0.6 (the most recent release at this writing) or later.
If you prefer, you can get netCDF from its home site, the Unidata Program Centre at the University Corporation for Atmospheric Research:
You will also need Perl 5, version 5.003 or later.
For easy ``one-stop shopping'', you can find Perl 5.004 (the latest version at this writing) at
Alternately, you can always find the latest version of Perl at at
any mirror of the Comprehensive Perl Archive Network (CPAN), in
src/5.0/. The easiest way is to go to:
If you prefer the old-fashioned way, here is a random selection of CPAN sites:
In addition to Perl itself, you will need the Getopt::Tabular module. This is also available at
or it can be found on any CPAN site, in
Getopt::Tabular, just unpack the archive and run the
following (you'll need write permission on the directories where Perl was
(If you can't write to the Perl installation directories (eg. if Perl was installed by the super-user, and you don't have root privileges) you can install the module to your home directory like this:
perl ./Makefile.PL PREFIX=~
but then you'll also need to add
~/lib/perl5/site_perl (or wherever
the module was actually installed) to your
The model used for stereotaxic registration is based on the 305 brain average MRI volume created at the Montreal Neurological Institute. Because of its size, this model is not included with the MNI_AutoReg package. But, it too is available from
Note that this archive only contains the files needed to build the full model; you will need to follow the instructions in the README file there to have the full set of model files. (See step 6 in the detailed instructions below.)
There are also two example MRI volumes available that you can use to test the package once you have everything built and installed. They are also both available from ftp://ftp.bic.mni.mcgill.ca/pub/mni_autoreg; the first (mni_autoreg_big_example_mri.mnc.gz) is a full-head, full-resolution ICBM T1-weighted scan. This file is about 23 MB uncompressed, so you may wish to consider using mni_autoreg_small_example_mri.mnc.gz (1.1 MB uncompressed), which is the same data but with the first preprocessing step (subsampling and cropping) performed on it. In order to check your results, you should also get mni_autoreg_example_results-0.98.tar.gz, which contains the output from running mritotal (the main user-level fitting program) on the two example volumes under various architectures.