vlisa_twosample.rst

LISA twosample test (vlisa_twosample)

The program vlisa_twosample performs a two-sample t-test on two groups of images including a correction for multiple comparisons using the LISA algorithm [2018_Lohmann]. For ease of use, the input images may be specified using wildcards as shown below. The output is a map thresholded such that FDR < alpha for every voxel. The default is alpha=0.05. The resulting image shows (1-FDR) so that larger values indicate higher significance.

The type of test can be specified using the option '-test'. The default ('-test pooled') is a twosample t-test based on pooled variance. The welch test ('-test welch') is a t-test applicable for unequal variances or unequal sample sizes. The paired test ('-test paired') can be used if the samples are paired. If wildcards are used to specify the input files, it is important to make sure that the pairs match.

Note that a region-of-interest mask is required. The mask should exclude non-brain voxels, and it may cover the entire brain. In the example below, the mask is in the file "braimmask.nii".

vlisa_twosample -in1 images1_*.v -in2 images2_*.v -mask brainmask.nii -out result.v -test welch

Note that this program also accepts input images in Nifti format (".nii" or ".nii.gz"), but the output is always in vista format. To convert the output to the Nifti format, use the following command:

vnifti -in result.v -out result.nii

Parameters of 'vlisa_twosample':

-help	Prints usage information.
-in1	Input files 1.
-in2	Input files 2.
-out	Output file.
-alpha	FDR significance level. Default: 0.05
-perm	Number of permutations. Default: 5000
-mask	Region of interest mask.
-test	Type of test to be performed [ pooled | paired | welch ]. Default: pooled
-seed	Seed for random number generation. Default: 99402622
-radius	Bilateral parameter (radius in voxels). Default: 2
-rvar	Bilateral parameter (radiometric). Default: 2.0
-svar	Bilateral parameter (spatial). Default: 2
-filteriterations
	Bilateral parameter (number of iterations). Default: 2
-cleanup	Whether to remove isolated voxels. Default: true
-j	Number of processors to use, '0' to use all. Default: 0

.. index:: lisa_twosample

References

[2018_Lohmann]

Lohmann G., Stelzer J., Lacosse E., Kumar V.J., Mueller K., Kuehn E., Grodd W., Scheffler K. (2018). LISA improves statistical analysis for fMRI. Nature Communications 9:4014. (link)