Guide to tutorial materials for ANTs/R
The goal of this tutorial is to use ANTs and ANTsR to turn raw data into clean data.
We will focus on pediatric brain magnetic resonance images available in the PTBP and show how to:
-
Organize and inspect the raw data and identify potential problem cases.
-
Identify hypotheses and quantification targets, given our data.
-
Develop a template construction procedure to address our hypotheses.
-
Attach prior distributions to the template.
-
Process the full data with ANTs pipelines.
-
Extend the baseline processing to interpret auxiliary modalities.
-
Perform univariate and multivariate statistical tests.
-
Explore prediction algorithms based on imaging data.
Along the way, we will cover theoretical foundations as well as practical insights arising from our experience with processing biological images numbering from the tens to tens of thousands.
Special sub-topics may include:
-
Baby brain
-
Multivariate neighborhood segmentation models
-
Longitudinal processing
-
Multivariate everything ...
-
ASL processing
-
Patch-based analyses for cross-modality inference ...
We will also entertain other requests.
Data sources
A complete multi-modality subject http://files.figshare.com/1701182/PEDS012_20131101.zip
PTBP demographics here
PTBP tutorial data [here](https://www.dropbox.com/s/5p0vlx8en9uzbge/ ants_tutorial_data.zip?dl=1)
Elderly subject data here
Software dependencies
The above tools depend on R, cmake and java.
Supported platforms include OSX and Linux variants.
These may work on windows but we do not regularly test there.