Analyzing DTI Using SPM5
Adapted from a protocol developed by Arianne Johnson (Elizabeth Sowell's group). Email or with questions.
To run voxel-based morphometric analyses on DTI data using SPM5, follow the instructions below:
Starting up SPM5
- Connect to cerebellum, making sure to forward your display using the -X option.
% ssh -X cerebellum.loni.ucla.edu
- Start up matlab by typing matlab at the command prompt. If matlab and spm are installed on your local computer, you may use these instead of logging in to cerebellum.
- Type spm pet in the matlab terminal to start up SPM5 in the PET/SPECT mode (although we're not looking at PET data, we are using 4D data, so this option is most appropriate).
Smoothing your images
- Click the "Smooth" button. Select your FA images for all of the subjects by double clicking in the windows to the right.
- Select smoothing width: 8x8x8.
- To test for significance between two groups, click the Basic models button. Under Design, select "Two-sample t-test".
- Enter the smoothed scans under the appropriate "Group 1" and "Group 2" subcategories. Leave the rest of the subcategories under "two-sample t-test" as is. Independence is assumed, unequal variance is assumed, grand mean scaling and ANCOVA are applicable to PET data.
- Select "Absolute" under "Masking". Enter a value of 0.2.
- Select "Directory ←X." Select the directory where you would like the output files to go. This must be a directory for which you have write permission.
- Select "Run".
- When this is complete, click the "Estimate" button. Choose the spm.mat file in the directory you just chose for output. Then click the "Results" button, and choose this spm.mat file again.
- Enter t-contrasts (for example, group 2>group1 would have values of -1 1. group 1>group 2 would be 1 -1).
- Select "no" on masking with other contrasts.
- Select FDR and enter a value of 0.05
- Use the "volume" and "overlays" options to investigate results.
Output Files and What They're For
- spmT_...hdr/img: Images of T-statistics.
- Mask.img: image indicating which voxels were included in the analysis.
- Con_0001, etc.: images of weighted parameter estimates.
- RPV.img: estimated resels (resolution elements) per voxel. This describes the actual spatial image resolution of the volume.
© 2009 LONI. All rights reserved.