FSL

Introduction

FSL is a library of useful neuroimaging software tools that we often use in our research. To use FSL on MASSIVE, simply load the module in a terminal with the command: module load fsl

You can also specify which specific version you wish to load: module load fsl/6.04

General Use

The FSL wiki has many tutorials for all the different tools within the library

Key tools for each modality include:

• fMRI - FEAT, MELODIC, FABBER, BASIL
• sMRI - BET, FAST, FIRST, FLIRT, FNIRT
• dMRI - FDT, TBSS, EDDY, TOPUP

Specific tools

ICA and fMRI de-noising

• FIX and AROMA are two useful tools developed by the FSL team for denoising fMRI data through ICA.

• These aren’t currently included in FSL because they have other dependencies, but they are installed as modules on MASSIVE and can be loaded accordingly

e.g., module load fix

For more detail on running FIX, look at ICA-FIX

MSM

• MSM is an important tool used in cortical registration

• MSM is included in FSL, but MSM_HOCR, a revised version of MSM, needs to be loaded as a separate module

FSLeyes

• FSLeyes is a visualisation software for FSL that can also be used for other neuroimaging data

• Like all visualisation tools on MASSIVE, this will require the addition of 'vglrun' at the start of your commands

• FSLeyes is required when inspecting FEAT/MELODIC ICA components for FSL-FIX

FSL FEAT

This is a tool that was previously used to minimally process fMRI data, but has largely been replaced with fmriPREP, it can still be useful to learn how early pre-processing steps are done and to gain an understanding of what goes on under the hood of fmriPREP

FSL Command Line Tools

FSL has a bunch of command line tools that can be helpful when processing MRI data, for an exhaustive list you can read through the FSL documentation, but i'll outline a few here that are commonly used.

• bet - this is a tool that removes the skull from an image (i.e. a Brain Extraction Tool) more information can be found here
• fsl_regfilt - this is the command used to regress information from an MRI image, it is the regression method used in ICA-AROMA and can be used to perform other forms of nuisance regression. There is no standalone wiki for this tool, but more information can be found here
• fslroi - this can be used to extract parts of an image. For instance, if you want to remove the first few frames from a functional image this can be done with fslroi
• fslmeants - this command extracts the mean ts across a series of voxels. For example if you wanted to get the average white matter signal from a functional image you would input the functional image and a white matter mask and you would get the mean white matter signal across the time series.
• fslmaths - this is a broad command that can achieve a lot of mathematical operations on an MRI image. There is no documentation on this that I am aware of, but you can get an idea of the functionality by typing the command without any inputs and it will give you a help description of what you can do.