normalize_mri
normalize MRI intensity values using a standard approach
normalize_mri [options] <input.mnc> <output.mnc>
DESCRIPTION
normalize_mri is a Perl script that normalizes MRI intensity values using N3 inhomogeneity correction and intensity standardization. The script performs a combined pipeline of non-uniformity correction followed by intensity normalization to produce a volume with standardized intensity values.
The normalization process includes:
- Non-uniformity correction using nu_correct to remove intensity bias
- Intensity standardization to match a reference model, ensuring consistent intensity values across different scans and scanners
This is useful for multi-subject studies where consistent intensity values are required for reliable tissue classification, voxel-based morphometry, or other quantitative analyses.
OPTIONS
-verbose- Print progress information during processing.
-clobber- Overwrite the output file if it already exists.
-model <file>- Specify the intensity model file to use as the normalization reference.
-mask <file>- Specify a brain mask to restrict the region used for intensity estimation.
EXAMPLES
Normalize a T1-weighted MRI volume:
normalize_mri t1_raw.mnc t1_normalized.mnc
Normalize with a specific model and brain mask:
normalize_mri -model /opt/models/icbm_t1.mnc -mask brain_mask.mnc t1_raw.mnc t1_normalized.mnc
Normalize with verbose output:
normalize_mri -verbose -clobber t1_raw.mnc t1_normalized.mnc
AUTHOR
Andrew Janke - McConnell Brain Imaging Centre, Montreal Neurological Institute, McGill University.
COPYRIGHTS
Copyright © 2000 by Andrew Janke