Tools for reproducible neuroimaging: an example from diffusion MRI

In this talk, I will discuss tools for reproducible research in neuroimaging. I will present a set of open-source libraries developed in the python programming language for reading/writing MRI data, for analysis of MRI data and for constructing reproducible analysis pipe-lines. I will demonstrate the use of these tools with a project focused on modeling the micro-structure of brain anatomy from diffusion-weighted magnetic resonance imaging (DW-MRI) data. I will present a framework for comparing different models of DW-MRI data by examining how well a model predicts an independent measurement (cross-validation).

eResearch NZ 2013 session type: 


Submitted by Tim McNamara on