FMRLAB Tutorial
v 2.0

©Jeng-Ren Duann & Scott Makeig, 2002
Swartz Center for Computational Neuroscience
Institute for Neural Computation
University of California San Diego

---

2. Installing FMRLAB

2.1 Download FMRLAB

The FMRLAB toolbox for fMRI data analysis using ICA can be downloaded from SCCN as a file named "fmrlab2.00.tar.gz". Under Microsoft Explorer, click the right mouse button and select "Save link as". Under Netscape, press SHIFT + left mouse button to download the compressed toolbox file and save it to disk.

2.2 Unzip and install FMRLAB

Copy the downloaded compressed file into an FMRLAB directory, for example, "/home/xxxx/matlab/" (Note: Throughout this tutorial, we will use this directory as the sample FMRLAB directory. Here, "xxxx" stands for any appropriate directory name). Use

      % tar xvfz fmrlab2.00.tar.gz

to uncompress the file. This will create a directory "fmrlab2.00" automatically under the directory you copied the file "fmrlab2.00.tar.gz" to and will save all the necessary files for running FMRLAB in the new FMRLAB directory

        /home/xxxx/matlab/fmrlab2.00

2.3 Add FMRLAB path to Matlab environment

Open the file "startup.m" using a text editor (if this file does not exist, create it). Add the line

       >> path(path,'/home/xxxx/matlab/fmrlab2.00');

to the end of file. Replace the "xxxx" here with actual pathname.

2.4 Edit FMRLAB settings file, "fmrlab_icadefs.m," to set ICA defaults

Open the file "fmrlab_icadefs.m" using a text editor. Find the following variables and change their contents accordingly.

      ICADIR = '/home/xxxx/matlab/fmrlab2.00/'
      ICABINARY = '/home/xxxx/matlab/fmrlab2.00/ica_linux'
      SC = '/home/xxxx/matlab/fmrlab2.00/binica.sc'
      FSLDIR = '/home/xxxx/matlab/fmrlab4.0/'

2.5 Download FMRLAB example dataset

The example data set used in this tutorial can be downloaded here. After successfully downloading the file, make a new directory (say, "/home/xxxx/matlab/example_data/") and copy the file to it. The example dataset contains two files, 2dseq_r1 and 2dseq_str. The first contains the functional images, the second the corresponding structural images (both in native image format). The functional images were acquired during a 5-minute experiment in which every 30 s the subject was shown brief 8-Hz flickering-checkerboard stimulus lasting 0.5 s. (See Duann et al., 2002 for details).

The image acquisition parameters for the functional images were:

• Image dimensions = 64 x 64 x 5
• FOV = 250 mm x 250 mm
• Slice thickness = 7 mm
• TR = 0.5 sec.
• Total number of scans = 610 (600 time points)
• Initial bad (dummy) scans = 10

The structural scans were T1-weighted images with the same slice positions, number of slices and FOV as the functional scans. However, they were acquired at 256 x 256 resolution to provide more structural detail than the functional scans. The structural image acquisition parameters were:

• Image dimensions = 256 x 256 x 5
• FOV = 250 mm x 250 mm
• Slice thickness = 7 mm

Back to the table of contents ...
Next chapter: Starting and quitting FMRLAB ...