MNI Automated Linear Registration PackageIn both diagnostic and research applications, the interpretation of magnetic resonance (MR) images of the human brain is facilitated when different data sets can be compared by visual inspection of equivalent anatomical planes. Quantitative analysis with pre-defined atlas templates often requires the initial alignment of atlas and image planes. Unfortunately, the axial planes acquired during separate scanning sessions are often different in their relative position and orientation, and these slices are not coplanar with those in the atlas.
A completely automatic method has been developed, based on multi-scale, three dimensional (3D) cross-correlation, to register a given volumetric data set to an average MRI brain (n > 300) aligned with the Talairach stereotaxic coordinate system. Once the data set is resampled by the transformation recovered by the algorithm, atlas slices can be directly super-imposed on the corresponding slices of the resampled volume (see below). The use of such a standardized space also allows the direct comparison, voxel-to-voxel, of two or more data sets brought into stereotaxic space.
A Perl script (mritotal) implements the multi-resolution fitting strategy that has been used to map more than 500 brains into stereotaxic space at the Montreal Neurological Institute. At the heart of this procedure is minctracc, the program that automatically finds the best linear transformation to map one volumetric data set (stored in MINC format, see below) on to another. The program uses optimization over a user selectable number of parameters to identify the best (according to a user-selected objective function) transformation mapping voxel values of the first data set into the second.
A paper describing the automatic linear stereotaxic registration procedure has appeared in:
D. L. Collins, P. Neelin, T. M. Peters and A. C. Evans, Automatic 3D Inter-Subject Registration of MR Volumetric Data in Standardized Talairach Space, Journal of Computer Assisted Tomography, 18(2) p192-205, 1994
The nonlinear registration procedure is described in:
DL Collins, CJ Holmes, TM Peters and AC Evans, Automatic 3D model-based neuro-anatomical segmentation Human Brain Mapping, 3(3) p 190-208, 1995
The install directions for this package are contained in the files INSTALL.quick and INSTALL. What follows here are instructions for installing prerequisite software.
You can obtain a copy of the latest version of the MNI_AutoReg software from:
http://packages.bic.mni.mcgill.ca/tgz
Look for a file named mni_autoreg-x.yy.tar.gz (where x.yy is 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
http://packages.bic.mni.mcgill.ca/tgz
If you prefer, you can get netCDF from its home site, the Unidata Program Centre at the University Corporation for Atmospheric Research:
ftp://unidata.ucar.edu/pub/netcdf
You will also need a recent version of Perl 5; Perl 5.002 or later is required.
You can always find the latest version of Perl at at any mirror of the Comprehensive Perl Archive Network (CPAN)
In addition to Perl itself, you will need the Getopt::Tabular module. This is also available at
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
http://pacakges.bic.mni.mcgill.ca/tgz/mni_autoreg_model-x.y.tar.gz
The version of mni_autoreg_model is NOT tied to the version of mni_autoreg. Choose the highest numbered version.
The primary scripts for registration are:
mritotal: the script that calls mincblur and minctracc to achieve stereotaxic registration
mritoself: intra-subject registration
Other scripts or programs installed are:
autocrop: for extracting and manipulating bounds of a MINC file
make_model: generate appropriate blurred and subsampled images required for `mritotal'
make_phantom: generate ellipsoid or rectangular phantom images
mincbbox: extract bounding box of image
mincblur: a program for volumetric convolution with a Gaussian blurring kernel
mincchamfer: compute a Chamfer distance transform of an image
minctracc: a program to estimate the transformation required for registration
xfmtool: perform miscellaneous operations on transform (.xfm) files
Some of these ship with manual pages, and all respond to `-help'. For further information, use the source, Luke! We'd be extremely please to receive documentation updates for any of the above!
We have gone to great lengths to ensure that MNI AutoReg compiles and runs correctly on a variety of Unix and Unix-like systems. However, we only have ready access to Linux, IRIX, and SunOS systems; successful use of the package on other systems is not necessarily assured. (For that matter, its successful use anywhere is not assured -- you get what you pay for! See the COPYING file.)
That said, if you have trouble building and/or testing the package, we'd like to hear about it. But just telling us "it didn't build" or "a test failed" is not enough -- ideally, you would provide us with a detailed breakdown of your system configuration (architecture, operating system and OS version, compiler and compiler version, C library -- including header files -- and its version), and a comprehensive explanation of what went wrong and how we can fix it without interfering with portability to any other system. We don't expect that every bug report will have this amount of information, but keep in mind that that's the ideal to strive for.
Please do not complain of portability problems unless you have first made sure that your C compiler and library (including header files!) are ANSI compliant.
Mritotal requires two external files to run. These are the configuration file (mritotal.cfg) and protocol file (default is mritotal.default.cfg, but you can specify others using the -protocol option). If you follow all the above instructions, then these files will be correctly customized and installed for your site. If things go wrong, though, you should know the following:
The directory where the configuration and protocol file(s) are installed defaults to /usr/local/etc/mni_autoreg. If you change this (ie. move the files after installing them) without rerunning "configure" and "make", then mritotal will not be able to find them. That's because "make" modifies mritotal so that it points to the directory for config files.
The location of the model files is specified in the configuration file. You should check that the directory where you installed the model files (which is specified in the Makefile for the mni_autoreg_model package) is the same as the directory given in mritotal.cfg (using the -modeldir option). For instance, if mni_autoreg_model/Makefile has the line:
MODELDIR = /data/avgbrain1/brain/images/modelthen mni_autoreg's perl/mritotal.cfg should have the line:
-modeldir /data/avgbrain1/brain/images/modelIf you find it necessary to make a new protocol for your site's data (for instance, if your MRI data consistently covers considerably more than the brain, but the heuristic used by the default protocol doesn't work -- see the mritotal man page for more information), then you should copy and change the default protocol file. For instance, you might want to call the protocol "mysite" -- in that case, copy mritotal.default.cfg to mritotal.mysite.cfg, and be sure to edit mritotal.cfg so that it uses the new protocol by default. That is, make sure it has "-protocol mysite" instead of "-protocol default".
That said, please notify Louis Collins louis@bic.mni.mcgill.ca if the heuristics used by the default protocol don't work for your data.
Good luck! Please let us know if anything goes wrong with the installation or testing:
Louis Collins Steve Robbins louis@bic.mni.mcgill.ca stever@bic.mni.mcgill.ca tel:(514)-398-1726,1996
McConnell Brain Imaging Center
NeuroImaging Laboratory
Montreal Neurological Institute
3801 University St.
Montreal, Quebec. H3A 2B4
Fax:(514)-398-8948