Elsevier

NeuroImage

Volume 62, Issue 2, 15 August 2012, Pages 782-790
NeuroImage

Review
FSL

https://doi.org/10.1016/j.neuroimage.2011.09.015Get rights and content

Abstract

FSL (the FMRIB Software Library) is a comprehensive library of analysis tools for functional, structural and diffusion MRI brain imaging data, written mainly by members of the Analysis Group, FMRIB, Oxford. For this NeuroImage special issue on “20 years of fMRI” we have been asked to write about the history, developments and current status of FSL. We also include some descriptions of parts of FSL that are not well covered in the existing literature. We hope that some of this content might be of interest to users of FSL, and also maybe to new research groups considering creating, releasing and supporting new software packages for brain image analysis.

Section snippets

A brief history of FSL

It all began in 1998 when the FMRIB Centre was officially opened, bringing Functional MRI to Oxford. Not only was the lab itself young, but so were the researchers in it, with almost everyone, and certainly all those in the Analysis Group, under 30; sadly, this is no longer true! We were all academic children, or grand-children, of Mike Brady, who was an instrumental player in setting up FMRIB alongside the original founders: Alan Cowey, George Radda and John Newsom-Davis. FMRIB consisted of

Some philosophy behind the software

We have discussed some aspects of the way that we program but not why we made these decisions. For instance, why did we choose to use C++ and scripts within a Unix-like environment? Although there are many factors, including the fact that we were already familiar with these, the principal reasons were speed, modularity, and portability. We wanted the software to be accessible to all, run quickly, and be powerful, by allowing maximum flexibility and adaptability.

It was (and arguably still is)

Past to present

It was June 2000 when we first released FSL, which at that time only contained a very small subset of the tools it contains today. Over time we have developed new methodology, and improved existing methodology, but also put significant efforts into support, documentation, publications and data formats. It now contains about 200,000 lines of code!

Future directions — XMI

The future for FSL is all about integration (see Fig. 4). We want to integrate more with tools like FreeSurfer, trying to make interoperation as seamless as possible. We also intend to add support for extra formats/dimensionalities: GIFTI (allowing surfaces to be represented), NIFTI2 (allowing 64-bit dimensions and storage), CIFTI (combining surface and volumetric components of GIFTI and NIFTI together, being developed within the NIH Human Connectome Project).

Scientifically, in the immediate

Acknowledgments

There are many people that have contributed to FSL and our research over the years and we would like to thank them all for what they have done, and apologise to any that we have forgotten. The list of developers (in approximately chronological order) is: YongYue “axe” Zhang, Peter Bannister, Dave Flitney, Heidi Johansen-Berg, Duncan Mortimer, Matthew Webster, Ivana Drobnjak, Rami Niazy, John Vickers, Tom Nichols, Natalie Voets, Jesper Andersson, Brian Patenaude, Gwenaëlle Douaud, Saad Jbabdi,

References (18)

There are more references available in the full text version of this article.

Cited by (0)

View full text