When you specify the sign, it changes the way the thresholding is done. If you use absolute (the default), then it ignores the sign. So if your threshold is 3, then anything above +3 and below -3 will survive, and so you end up with more and bigger clusters.

doug



Dan Dillon wrote:

Dear FreeSurfers,

 

I’m trying to replicate some functional analyses that were originally done in SPM2, and am using mri_volcluster. Here’s my command: mri_volcluster --in Old8/bold/taumax30_2_21/tal-ffx/RSF_v_NIF/sig --in_type bfloat --reg Old8/bold/taumax30_2_21/tal-ffx/register.dat --thmin 3 --minsizevox 8.

 

I’ve noticed that the size (in voxels) and number of clusters changes if I play with the –sign flag. I am confused by this.

 

Here’s the output not using the –sign flag:

 

Cluster   Size(n)   Size(mm^3)     TalX   TalY    TalZ              Max

  1          28       1792.0     -19.80  -96.70    8.53           5.28907

  2           8        512.0      39.60   26.79   -8.07           4.15380

  3           9        576.0      99.00   16.69  -54.67           4.09661

  4          10        640.0      87.12   63.11   18.95          -3.82155

  5          13        832.0      67.32  -42.44    5.81           3.79067

  6          11        704.0     -39.60  -16.43   59.78          -3.62503

  7          11        704.0     -87.12  -75.98  -43.30          -3.59305

  8          73       4672.0      43.56  -22.33   19.54          -3.43893

  9          18       1152.0      43.56  -44.81  -41.49          -3.16065

 10           8        512.0     -51.48  -32.30   53.20          -3.09981

 11          15        960.0     -63.36  -20.49   56.30          -3.00613

 12          48       3072.0     -71.28   -7.20   11.41          -3.00547

 

And here’s the output with sign set to positive:

 

Cluster   Size(n)   Size(mm^3)     TalX   TalY    TalZ              Max

  1          35       2240.0     -19.80  -96.70    8.53           5.28907

  2          21       1344.0      39.60   26.79   -8.07           4.15380

  3          13        832.0      99.00   16.69  -54.67           4.09661

  4          10        640.0     -71.28   33.20  -35.31           3.91335

  5          21       1344.0      67.32  -42.44    5.81           3.79067

  6           9        576.0      35.64  -88.58   15.49           3.63267

 

Why does the size of the first cluster change? Also, where is cluster 4 from the second output in the first output? Is there some re-calculation going on here?

 

Dan Dillon, Ph.D.

Post-doctoral Fellow

Affective Neuroscience Lab

Dept. of Psychology, Harvard University

Phone: (617) 496-5222

E-mail: dillon@wjh.harvard.edu

 


_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

-- 
Douglas N. Greve, Ph.D.
MGH-NMR Center
greve@nmr.mgh.harvard.edu
Phone Number: 617-724-2358 
Fax: 617-726-7422

In order to help us help you, please follow the steps in:
surfer.nmr.mgh.harvard.edu/fswiki/BugReporting