PNI is actively involved in the development and
testing of new analytic tools and methods for analyzing neuroimaging
data. Below are some of our active development projects. More information,
reliability and validity, as well as some sample data will be available
soon.
BRAINSTRIP
BRAINSTRIP is a semi-automated brain extraction
tool, implemented as a plug-in to MIPAV. The main developers of
this tool are Pierre-Louis Bazin and Dzung Pham of the MedIC
lab. The program implements the algorithm by Goldszal
at al. (1998). PNI is involved in the testing of this software,
as well as establishing the reliability and validity of the method.
The plug-in algorithm and evaluations are currently in press. For
testing, you can download the software from the MedIC Site.
GRAT (Group ROI
Analysis Toolbox)
The GRAT is under active development by Michael
Yassa. It is a plug-in to SPM that utilizes the MarsBaR routines
to extract region of interest time course for groups of subjects,
calculates average time courses and percent voxels activated, as
well as mean and maximum amplitudes. It also includes plotting routines
to co-plot time-course data with task parameters. This project is
still under construction.
ABSEG (Atlas-Based
Segmentation)
ABSEG was developed by Adam Dziorny as a plug-in
to MIPAV, re-implementing an algorithm for whole brain segmentation
based on Atlas priors by Bruce Fischl at MGH. The software provides
a tool for the automated labeling of neuroanatomical structures
in the human brain. This algorithm combines spatial and intensity
information to improve registration to manually created atlases.
This tool is still undergoing testing. For more information about
this tool, please contact Scott Yeagy.
DEMONS
DEMONS was originally developed by Thirion et
al. (1998) as an image registration method using optical flow techniques.
It was re-implemented by the MedIC
lab as a method to assess longitudinal atrophy in patients at
increased risk for developing AD. The use of DEMONS as a method
to reliably assess these differences is still undergoing tesing.
DEMONS is also implemented as a plug-in to MIPAV.
VOXEL-BASED TECHNIQUES
With the onset of voxel-based morphometric analyses
a few years ago, several applications of these methods have become
possible. PNI is developing new methods to extend the use of VBM
and VBM-like methods to the assessment of differences across groups
in different imaging modalities, such as diffusion weighted imaging,
and fractional anisotropy images. In addition, we are also working
on ways to conduct these voxel-based analyses with regional specificity
to improve the quality of the registration in the hypothesized areas
of interest, and increase statistical power.
