bah2016_registration

Registration method for in situ hybridization images to fMRI standard coordinated brain (e.g. Waxholm Brain Atlas). This program was parallelized for PC clusters to increace registration speed.

Developed by Brain Atlas Hackathon Registration Team (https://www.neuroinf.jp/bah2015/?ml_lang=en)

Dependency

ipython
ipyparallel
ImageJ/Fiji
ITK

Files and Directories

Parallel registration input files and dirs
- ImageRegistration9: registration program from ITK
- filelist.txt: write down all TIF file name of in situ hybridization images
- work3: Monochromed (and resized) TIF files of in situ hybridization images
- SB_bin: resliced standard brain files (from Waxholm Space)
Parallel registration output files and dirs
- work4: results of registration for all standard brain slices
- log4: registration log
- metric: metric value for each slice
- metric_fig: graph of metric value

usage

Parallelized registration with itk and ipyparallel

start ipyparallel controller

$ ipcontroller --ip=*

start ipython engine by SGE (MPI)

$ qsub start_engine.sh

start registration job

$ cd registration
$ python ParaRegistration.py

neuroinformatics/bah2016_registration