Download the Package and Tutorial


iBEAT (previously called LIBRA) is a toolbox with graphical user interfaces for processing infant brain MR images. Longitudinal (or single-time-point) multimodality (including T1, T2, and FA) (or single-modality) data can be processed using the toolbox. Main functions of the software (step by step) include image preprocessing, brain extraction, tissue segmentation and brain labeling. Linux operating system is required for the stand-alone executable software. The graphical user interfaces and overall framework of the iBEAT software are implemented in MATLAB. The image processing functions are implemented with the combination of C/C++, MATLAB, Perl and Shell languages. Parallelization technologies are used in the software to speed up image processing. More details can be found in the Tutorial.

Fig. 1 Graphical User Interfaces. The left is the main GUI of the iBEAT software. The right is the GUI of the image preprocessing module.


  • Linux operating system is required.
  • Installation Steps:
    1. Download the ibeat package and unzip the (e.g., the package is unzipped in /home/programs/ibeat).
    2. Setup environment for the ibeat as follows:
      1. Edit the shell resource file in the home directory of the user (cd ~) :
        1. Add the following two lines in the .cshrc file if csh/tcsh is used
          • setenv IBEAT_HOME /home/programs/ibeat
          • source $IBEAT_HOME/ibeatEnv.csh
        2. Add the following two lines in the .bashrc file if bash is used
          • export IBEAT_HOME=/home/programs/ibeat
          • source $IBEAT_HOME/ibeatEnv.bash
      2. Restart the shell to update the environment.
    3. Use command ibeat to start the software.
  • Refer to the Tutorial to use the software.


The processing pipeline of infant brain MR images implemented in the iBEAT software is shown in Fig. 2. The brain extraction function uses the method proposed by Shi et al [1]. The tissue segmentation function uses the method proposed by Wang et al [2,3]. The brain labeling uses the HAMMER registration method proposed by Shen et al [4].

Fig. 2 The processing pipeline of infant brain MR images implemented in the iBEAT software.


  1. Feng Shi, Li Wang, John H. Gilmore, Weili Lin and Dinggang Shen. Learning-based Meta-Algorithm for MRI Brain Extraction. MICCAI 2011, Toronto, Canada, Sep. 18-22, 2011.
  2. Li Wang, Feng Shi, Pew Thian Yap, John H Gilmore, Weili Lin and Dinggang Shen. Accurate and Consistent 4D Segmentation of Serial Infant Brain MR Images. Multimodal Brain Image Analysis (MBIA 2011), Toronto, Canada, Sep. 18, 2011.
  3. Li Wang, Feng Shi, Pew-Thian Yap, Weili Lin, John H. Gilmore, Dinggang Shen. Longitudinally guided level sets for consistent tissue segmentation of neonates. Hum Brain Mapp. 2011.
  4. Dinggang Shen and Christos Davatzikos. HAMMER: Hierarchical Attribute Matching Mechanism for Elastic Registration. IEEE Transactions on Medical Imaging, 21(11):1421-1439, 2002.