FSL Cluster Analysis (cluster)¶
Library: FSL | Docker Image: brainlife/fsl
Function¶
Identifies contiguous clusters of suprathreshold voxels in statistical images and reports their properties.
Modality: Statistical map (z-stat or p-value 3D NIfTI).
Typical Use: Cluster-based thresholding and extracting peak coordinates from statistical maps.
Key Parameters¶
-i (input stat image), -t (z threshold), -p (p threshold), --oindex (cluster index output), --olmax (local maxima output), -c (cope image for effect sizes)
Key Points¶
Reports cluster size, peak coordinates, and p-values. Use with -c to get mean COPE within clusters. GRF-based p-values require smoothness estimates.
Inputs¶
| Name | Type | Required | Label | Flag |
|---|---|---|---|---|
input |
File |
Yes | Input statistical image (e.g., z-stat) | --in= |
threshold |
double |
Yes | Threshold value for cluster formation | --thresh= |
oindex |
string |
No | Output cluster index image filename | --oindex= |
othresh |
string |
No | Output thresholded image filename | --othresh= |
olmax |
string |
No | Output local maxima text file | --olmax= |
olmaxim |
string |
No | Output local maxima image filename | --olmaxim= |
osize |
string |
No | Output cluster size image filename | --osize= |
omax |
string |
No | Output max intensity image filename | --omax= |
omean |
string |
No | Output mean intensity image filename | --omean= |
opvals |
string |
No | Output log p-values image filename | --opvals= |
pthresh |
double |
No | P-threshold for clusters | --pthresh= |
dlh |
double |
No | Smoothness estimate (sqrt determinant of Lambda) | --dlh= |
volume |
int |
No | Number of voxels in mask | --volume= |
cope |
File |
No | COPE image for effect size reporting | --cope= |
peakdist |
double |
No | Minimum distance between peaks in mm | --peakdist= |
connectivity |
int |
No | Voxel connectivity (6, 18, or 26) | --connectivity= |
fractional |
boolean |
No | Interpret threshold as fraction of robust range | --fractional |
mm |
boolean |
No | Use mm coordinates instead of voxel | --mm |
find_minima |
boolean |
No | Find minima instead of maxima | --min |
num_maxima |
int |
No | Number of local maxima to report | --num= |
xfm |
File |
No | Linear transformation matrix file | --xfm= |
stdvol |
File |
No | Standard space volume for coordinate transformation | --stdvol= |
warpvol |
File |
No | Warp field for non-linear transformation | --warpvol= |
minclustersize |
boolean |
No | Print minimum significant cluster size | --minclustersize |
no_table |
boolean |
No | Suppress printing of cluster table | --no_table |
verbose |
boolean |
No | Enable verbose output | --verbose |
Accepted Input Extensions¶
- input:
.nii,.nii.gz - cope:
.nii,.nii.gz - xfm:
.mat - stdvol:
.nii,.nii.gz - warpvol:
.nii,.nii.gz
Outputs¶
| Name | Type | Glob Pattern |
|---|---|---|
cluster_table |
File |
cluster_table.txt |
cluster_index |
File |
$(inputs.oindex).nii.gz, $(inputs.oindex).nii |
thresholded_image |
File |
$(inputs.othresh).nii.gz, $(inputs.othresh).nii |
local_maxima_txt |
File |
$(inputs.olmax) |
local_maxima_image |
File |
$(inputs.olmaxim).nii.gz, $(inputs.olmaxim).nii |
size_image |
File |
$(inputs.osize).nii.gz, $(inputs.osize).nii |
max_image |
File |
$(inputs.omax).nii.gz, $(inputs.omax).nii |
mean_image |
File |
$(inputs.omean).nii.gz, $(inputs.omean).nii |
pvals_image |
File |
$(inputs.opvals).nii.gz, $(inputs.opvals).nii |
log |
File |
cluster.log |
Output Extensions¶
- cluster_table:
.txt - cluster_index:
.nii,.nii.gz - thresholded_image:
.nii,.nii.gz - local_maxima_txt:
.txt - local_maxima_image:
.nii,.nii.gz - size_image:
.nii,.nii.gz - max_image:
.nii,.nii.gz - mean_image:
.nii,.nii.gz - pvals_image:
.nii,.nii.gz
Docker Tags¶
Available versions: latest, 6.0.4-patched2, 6.0.4-patched, 6.0.4, 6.0.4-xenial, 5.0.11, 6.0.0, 6.0.1, 5.0.9
Categories¶
- Utilities > FSL > Clustering