DESCRIPTION

FSL is a comprehensive toolkit for the analysis of medical brain imaging data. It provides tools for data preprocessing, brain extraction, image registration as well as a broad spectrum of statistical analyses. FSL provides an easy to use GUI.

FSL requires a number of environment variables to be set properly. To achieve this, please source the config script from /etc/fsl/4.1/fsl.sh. One way to do this automatically is to add a line containing

 . /etc/fsl/4.1/fsl.sh

to the .bashrc file in a users home directory (or a corresponding file for other shells than bash). You may need to create such a file first.

Alternatively, the Debian package installs symlinks for all FSL tools in /usr/bin that take care of loading the appropriate configuration automatically. The symlink names are prefixed with the corresponding FSL version to prevent filename conflicts and allow multiple version to coexist on the same system. For example, the slicer command is available as fsl#FSLMVERSION-slicer. Please note that for using FSL tools via these symlinks it is not necessary to perform the setup procedure describe above. However, this convenience feature is Debian-specific. Scripts that employ FSL commands and are intended to be shared with others should rely on unprefixed command names and the FSL environment variable setup.

The FSL user interface sometimes refers to an FSLDIR. On Debian systems this is /usr/share/fsl/4.1 for FSL version 4.1.

FSL documentation and online help is provided by the fsl-doc-4.1 package.

Questions regarding the usage of FSL or related theory can be posted on the FSL mailing list (see http://www.jiscmail.ac.uk/lists/fsl.html).

PARALLELIZATION

FSL can be configured to submit jobs to a running Sun Gridengine (SGE). While not all functionality implemented in FSL supports this kind of parallelization, the most demanding processing tools (e.g. POSSUM and FDT) do.

FSL should be able to use SGE even if the SGE_ROOT environment variable is not set, as in the case of the Debian package of SGE. To enable batch processing simply set

FSLPARALLEL=1

in the shell where FSL is running. This can be enabled permanently in

/etc/fsl/4.1/fsl.sh

or on a case by case basis.

FILES

/etc/fsl/4.1/fsl.sh

Main FSL configuration

/etc/fsl/4.1/feat.tcl

Default settings for FEAT

AUTHOR

FSL was written by FMRIB, Oxford, UK.

This manual page was written by Michael Hanke <[email protected]>, for the Debian project (but may be used by others).