This repository contains materials for an introductory workshop on fMRI preprocessing using fMRIPrep. The goal of this workshop is to provide a practical, hands-on introduction to reproducible preprocessing of fMRI data using BIDS-formatted datasets and containerized workflows.
Authors: Fabian Dorok and Sterre van de Langenberg
During this workshop, participants will learn:
- The BIDS data structure and BIDS conversion using Heudiconv
- How to run fMRIPrep using Docker (and WSL installation for Windows Users)
- Understanding fMRIPrep outputs
- Quality control of fMRI preprocessing
- Basics of reproducible neuroimaging workflows
To run the pipeline locally, you will need:
- Docker installed
- FreeSurfer license file
- A BIDS-formatted dataset
- At least 8–16 GB RAM (recommended: 32 GB+)
Example datasets (defaced) used in this workshop are stored in the Datasets/ folder.
Unzip to your local machine before running HeuDiConv (BIDS conversion).
Data can also be found here: https://drive.google.com/file/d/1niOEgwtN6ZeyWhGzxSUMYGn_NtEUdrTl/view?usp=drive_link
We created a Jupyter notebook for coding along at home (as a reference after the workshop).
We also uploaded a script on how to mount network drives in WSL (Scripts/ folder).