HALFpipe: Advancing reproducible fMRI analysis based on fMRIPrep

[HippocampusGirl]

Title

HALFpipe: Advancing reproducible fMRI analysis based on fMRIPrep

Short description and the goals for the OHBM BrainHack

Using standardized pipelines leads to consistent results, right? We ran fMRIPrep 20.2.7 one hundred times and then calculated functional connectivity matrices using HALFpipe.

1. We did this in a multiverse approach where we varied which denoising strategy was applied before connectivity matrix calculation. An interesting next step here is to figure out how to quantify the variability in these matrices depending on the strategy. Are there any strategies that better than others?
   Any interested contributors may download the data and attempt to answer this question with us!
2. We would like to repeat this procedure with newer versions of fMRIPrep, and compare to the baseline that we already ran. How does the variability between versions compare to the variability within a single version?
   This will require some programming to load in derivatives files from different versions of fMRIPrep and then apply the Nipype workflows from HALFpipe to them. We are welcoming contributors willing to expand their familiarity with Nipype workflows in Python or BIDS derivatives.
3. The Enhancing Neuro Imaging Genetics through Meta Analysis (ENIGMA) Consortium conducts the largest brain imaging studies in the world, involving over 500 institutions in 45 countries worldwide. At this time, the ENIGMA Consortium has processed more than fifty thousand fMRI scans using fMRIPrep, mostly using HALFpipe as a user interface to fMRIPrep for ease of use. HALFpipe also calculates downstream results such as functional connectivity maps and matrices.
   While fMRIPrep supports resampling fMRI data to the cortical surface, the current HALFpipe workflows currently can not use those outputs. We would like to change this, and have a roadmap of the Nipype workflows in HALFpipe that need to be adapted.
   We would love to discuss the best approaches and potentially even try implementing some of the changes.

Link to the Project

https://github.com/HALFpipe/HALFpipe

Image/Logo for the OHBM brainhack website

Project lead

Name	GitHub	Discord
Violeta Céspedes	@lalalavi	@lalalaaaavi
Lea Waller	@HippocampusGirl	@wwwlea

Main Hub

Seoul

Link to the Project pitch

No response

Other hubs covered by the leaders

• Seoul
• Hybrid (Asia / Pacific)
• Hybrid (Europe / Middle East / Africa)
• Hybrid (Americas)

Skills

Necessary	Not necessary
Python	Nipype Brain Imaging Data Structure (BIDS)

Recommended tutorials for new contributors

No response

Good first issues

No response

Twitter summary

No response

Short name for the Discord chat channel (~15 chars)

halfpipe

Please read and follow the OHBM Code of Conduct

• I agree to follow the OHBM Code of Conduct during the hackathon