Skip to content

Brain Imaging Center Image Processing Pipelines

Notifications You must be signed in to change notification settings

jadenecke/bic-pipelines

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

44 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Brain Imaging Center Image Processing Pipelines 

Introduction
------------

This project includes basic image processing pipelines intended 
to be run on raw output from the scanner (in Minc format) and produce output 
suitable for population analysis in standard stereotaxic space.


Installation
-------------
All pipelines are written in high-level scripting language (Perl,Python or Ruby) 
and require complete installation of minc-toolkit  ( https://github.com/BIC-MNI/minc-toolkit )

installation consist of copying all scripts into the same directory, and setting PATH and PERL5LIB to include this directiry

MNI_DATADIR should point to the location of icbm152_model_09c ( or specify location of the models in the command line)


Usage
--------------
Pipeline processes a dataset with one T1w scan and (optionally) one T2w and PDw scans acquired using
dual echo sequnce (i.e t2w and pdw scans are expected to be co-registered)


to start:

standard_pipeline.pl <subject id> <visit id> \
 <path to T1w scan> \
 [path to T2w scan] \
 [path to T2w scan] \
 [ --prefix <output prefix> ]
 [ --model_dir <location of anatomical models> ]
 [ --model <name of the anatomical model (default mni_icbm152_t1_tal_nlin_sym_09c) ]
 
Will create subdirectory structure in 
<output prefix> :

<ID> specifies subject ID ( any sequence of characters, without spaces), <Visit> - visit label (for example V1 V2 etc.)
<T1 file> - file name of T1 scan
[T2 file] - optional T2 scan
[PD file] - optional PD scan
<directory> - base directory where all results will be placed. The pipeline script will create subdirectory <directory>/<ID>/<Visit>/ ....
and will put all files there:

./clp - intensity normalization (map all values to the 0-100 range approxiately)
./clp/clamp_<ID>_<Visit>_t1w.mnc t1w scan
./clp/clamp_<ID>_<Visit>_t2w.mnc t2w scan (if present)
./clp/clamp_<ID>_<Visit>_pdw.mnc pdw scan (if present)

./tal - registration in stereotaxic space
./tal/tal_xfm_<ID>_<Visit>_t1w.xfm - registration parameters for t1w
./tal/tal_xfm_<ID>_<Visit>_t2w.xfm - registration parameters for t2w/pdw (if present)
./tal/tal_<ID>_<Visit>_t1w.mnc  - t1w scan in the stereotaxic space
./tal/tal_<ID>_<Visit>_t2w.mnc  - t2w scan in the stereotaxic space
./tal/tal_<ID>_<Visit>_pdw.mnc  - pdw scan in the stereotaxic space
./tal/tal_noscale_xfm_<ID>_<Visit>_t1w.xfm - registration parameters fot t1w without scaling
./tal/tal_noscale_xfm_<ID>_<Visit>_t2w.xfm - registration parameters fot t2w without scaling
./tal/tal_noscale_xfm_<ID>_<Visit>_pdw.xfm - registration parameters fot pdw without scaling
./tal/tal_noscale_<ID>_<Visit>_t1w.mnc - t1w scan in stereotaxic space, unscaled
./tal/tal_noscale_<ID>_<Visit>_t2w.mnc - t1w scan in stereotaxic space, unscaled
./tal/tal_noscale_<ID>_<Visit>_pdw.mnc - t1w scan in stereotaxic space, unscaled
./tal/tal_comp_msk_<ID>_<Visit>.mnc - brain mask 

./nl - nonlinear registration into sterotaxic space
./nl/nl_xfm_<ID>_<Visit>.xfm - registration parameters
./nl/nl_xfm_<ID>_<Visit>_grid_0.mnc
./nl/nl_<ID>_<Visit>_t1w.mnc - scan nonlinearly registered into stereotaxic space
./nl/nl_jacobian_<ID>_<Visit>.mnc - jacobian determinant map of the nonlinear part of transformation

./tal_cls  - tissue classification using linear registration into sterotaxic space
./tal_cls/tal_clean_<ID>_<Visit>.mnc - tissue map

./lob  - brain segmentation
./lob/tal_clean_lob_<ID>_<Visit>.mnc - brain segmentation using linear registration 
./lob/nl_clean_lob_<ID>_<Visit>.mnc  - brain segmentation using nonlinear registration (more precise)

./qc  - Quality Control Images
./qc/qc_face_<ID>_<Visit>.jpg - face rendering
./qc/qc_t1w_<ID>_<Visit>.jpg  - head with stx target outline (linear registration)
./qc/qc_tal_t1t2_<ID>_<Visit>.jpg - t2w and t1w co-registration
./qc/qc_nl_reg_<ID>_<Visit>.jpg - brain with stx target outline (nonlinear registration)
./qc/qc_t1w_lc_clean_<ID>_<Visit>.jpg - tissue classification (linear)
./qc/qc_nl_t1w_segment_lobe_<ID>_<Visit>.jpg - brain segmentation (nonlinear)

./vol
./vol/nl_<ID>_<Visit>.txt - text file with volumes of various substructures (in native space)

About

Brain Imaging Center Image Processing Pipelines

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Perl 99.5%
  • CMake 0.5%