~/projects/fmri-preprocessing $ cat README.md

Reproducible fMRI preprocessing

Getting brain scans ready for analysis, in a way that can be repeated.

context
EAT Lab, Maastricht University
when
2025
status
Finished · internship contribution
  • BrainVoyager
  • FSL
  • Linux
  • Conda / Micromamba
  • ANTs setup
  • Freesurfer setup

// the problem

Before you can analyse fMRI data, every run has to be corrected for head motion and slice timing, then aligned. By hand that's slow and hard to repeat exactly.

// what I built

During my research internship at the EAT Lab in Maastricht, I set up and tested a Linux environment for the lab's neuroimaging tools (FSL, ANTs, Freesurfer, managed with Conda/Micromamba) and processed task runs with BrainVoyager and FSL. I helped make the lab's workflow more automated; the pipeline was a team effort.

// how it works

The standard steps every run goes through:

// result

Task runs prepared for analysis, and a tested environment for the lab's tools.

// what I took away

Reproducible research starts with the software setup. If nobody can rebuild that, nobody can rebuild the results either.