Open Workflows (Lightning talk): dMRIPrep - A robust and reproducible pipeline for dMRI data pre-processing

[jsheunis]

dMRIPrep - A robust and reproducible pipeline for dMRI data pre-processing

By Michael Joseph, The Centre for Addiction and Mental Health, Toronto, Canada

• Theme: Open Workflows
• Format: Lightning talk

Abstract

dMRIPrep is an analysis-agnostic pipeline that seeks to address the challenge of robust and reproducible pre-processing for whole-brain dMRI data. The pipeline is based on Nipype and encompasses a large set of tools from other neuroimaging packages. It performs basic preprocessing steps such as head-motion correction, susceptibility-derived distortion correction, eddy current correction, etc., providing outputs that can be easily submitted to a variety of diffusion models. This pipeline was designed to provide the best software implementation for each stage of preprocessing and will be continually updated as better methods become available. The Open Science Room is an excellent venue for dMRI enthusiasts to meet. We would like to demo dMRIPrep's core functionality and seek testing, feedback, and engage contributors from the neuroimaging community. This talk will contain an interactive overview of dMRI data pre-processing with dMRIPrep, as well as highlight some novel methods and quality control visualizations being developed.

Collaborators: https://github.com/nipreps/dmriprep/graphs/contributors Adam Richie-Halford Anisha Keshavan Michael Joseph Garikoitz Lerma-Usabiaga Derek Pisner Matthew Cieslak Erin Dickie Jelle Veraart Ross Lawrence Ariel Rokem Oscar Esteban

Useful Links

https://github.com/nipreps/dmriprep https://github.com/nipreps/dmriprep/graphs/contributors

Tagging @josephmje

[josephmje]

I am happy with the details here to include in the program

[josephmje]

Slides can be found here