/PICSL_MALF

Updated code from project https://www.nitrc.org/projects/picsl_malf/ Imported from: https://www.nitrc.org/frs/?group_id=634

Primary LanguageC++

PICSL Multi-Atlas Segmentation Tool

This package contains source code for joint label fusion [1] and corrective learning [2], which were 
applied in MICCAI 2012 Grand Challenge on Multi-Atlas Labeling and finished in the first place.

This software is provided for research purpose only under the GNU General Public License. 

Joint label fusion is for combining candidate segmentations produced by registering and warping 
multiple atlases for a target image. Corrective learning can be applied to further reduce 
systematic errors produced by joint label fusion (see [2] for detail). In general, corrective 
learning can be applied to correct systematic errors produced by other segmentation methods 
as well. If you use this software to produce results for a publication, please cite the 
following paper(s) accordingly.

[1] H. Wang, J. W. Suh, S. Das, J. Pluta, C. Craige, P. Yushkevich, "Multi-atlas segmentation 
with joint label fusion," IEEE Trans. on Pattern Analysis and Machine Intelligence, 35(3), 611-623, 2013

[2] H. Wang, S. R. Das, J. W. Suh, M. Altinay, J. Pluta, C. Craige, B. B. Avants, and P. A. Yushkevich,
"A Learning-Based Wrapper Method to Correct Systematic Errors in Automatic Image Segmentation:
Consistently Improved Performance in Hippocampus, Cortex and Brain," Neuroimage, vol. 55, iss. 3,
pp. 968-985, 2011.


INSTRUCTIONS FOR COMPILING THE CODE

A cmake file is provided to facilitate compiling the code. The user needs to set up configurations for
compiling. To do so, go to the folder of the software package. Then type

ccmake .

to setup the enviroment. After this is done, type

make

to complie and build the executable files. Our program uses ITK's I/O functions to handle image input 
and output. Hence, it requries prebuilt ITK. The following three executable files will be built, 

jointfusion : joint label fusion
bl          : learning classifiers for correcting systematic errors
sa          : apply the learned classifiers to correct systematic segmentation errors on testing images

Type each command to see details on how to use them. 


09/04/2012
Hongzhi Wang
wanghongzhi78@gmail.com