AFNI 3D Intensity Uniformization (3dUnifize)¶
Library: AFNI | Docker Image: brainlife/afni
Function¶
Corrects intensity inhomogeneity (bias field) to produce uniform white matter intensity.
Modality: T1-weighted or T2-weighted 3D NIfTI/AFNI volume.
Typical Use: Bias correction before segmentation or registration.
Key Parameters¶
-input (input), -prefix (output), -T2 (for T2-weighted input), -GM (also unifize gray matter)
Key Points¶
Fast bias correction alternative to N4. Works well for T1 images by default. Use -T2 flag for T2-weighted images.
Inputs¶
| Name | Type | Required | Label | Flag |
|---|---|---|---|---|
input |
File |
Yes | Input dataset | -input |
prefix |
string |
Yes | Output dataset prefix | -prefix |
T2 |
boolean |
No | Process input as T2-weighted (inverts contrast) | -T2 |
EPI |
boolean |
No | Process T2/T2* weighted EPI time series | -EPI |
GM |
boolean |
No | Scale to unifize gray matter intensities | -GM |
Urad |
double |
No | Radius in voxels for processing ball (default 18.3) | -Urad |
clfrac |
double |
No | Automask clip level fraction (0.1-0.9, default 0.2) | -clfrac |
noduplo |
boolean |
No | Disable half-size volume processing step | -noduplo |
nosquash |
boolean |
No | Disable reduction of large intensity values | -nosquash |
T2up |
double |
No | Upper percentile for T2-T1 inversion (90-100) | -T2up |
rbt |
string |
No | Algorithm parameters (radius bottom_percentile top_percentile) | -rbt |
ssave |
string |
No | Save white matter scale factors to dataset | -ssave |
amsave |
string |
No | Save automask-ed input dataset | -amsave |
quiet |
boolean |
No | Suppress progress messages | -quiet |
Accepted Input Extensions¶
- input:
.nii,.nii.gz,+orig.HEAD,+orig.BRIK,+tlrc.HEAD,+tlrc.BRIK
Outputs¶
| Name | Type | Glob Pattern |
|---|---|---|
unifized |
File |
$(inputs.prefix)+orig.HEAD, $(inputs.prefix)+tlrc.HEAD |
scale_factors |
File |
$(inputs.ssave)+orig.HEAD, $(inputs.ssave)+tlrc.HEAD |
automask |
File |
$(inputs.amsave)+orig.HEAD, $(inputs.amsave)+tlrc.HEAD |
log |
File |
$(inputs.prefix).log |
Output Extensions¶
- unifized:
+orig.HEAD,+orig.BRIK,+tlrc.HEAD,+tlrc.BRIK - scale_factors:
+orig.HEAD,+orig.BRIK,+tlrc.HEAD,+tlrc.BRIK - automask:
+orig.HEAD,+orig.BRIK,+tlrc.HEAD,+tlrc.BRIK
Docker Tags¶
Available versions: latest, 16.3.0
Categories¶
- Structural MRI > AFNI > Bias Correction