confounds: deconfounding library to properly handle confounds

[raamana]

Title

confounds: deconfounding library to properly handle confounds

Short description and the goals for the OHBM BrainHack

Develop a python library of methods to handle confounds in various neuroscientific analyses, esp. statistics and predictive modeling. More info and slides here: https://crossinvalidation.com/2020/03/04/conquering-confounds-and-covariates-in-machine-learning/

Link to the Project

https://github.com/raamana/confounds

Image for the OHBM brainhack website

No response

Project lead

Pradeep Reddy Raamana @raamana

Main Hub

Glasgow

Other Hub covered by the leaders

• [ ] Glasgow
• [ ] Asia / Pacific
• [ ] Europe / Middle East / Africa
• [ ] Americas

Skills

• python programming (preferably intermediate or better, but can work with basic skills)
• some statistics
• documentation ability

Recommended tutorials for new contributors

• [ ] Set up a brainhack friendly computing environment
• [ ] A introduction to Bash
• [ ] Python: Writing a script
• [ ] Python: Data analysis with Python
• [ ] Python: Visualisation
• [ ] Machine learning basics
• [ ] BIDS
• [ ] Machine learning for neuroimaging
• [X] VCS: Using Git and Github
• [ ] VCS: Data management with Datalad
• [ ] Virtualization
• [ ] High Performance Computing: Oracle cloud resource

Good first issues

• Add a simple test to target a specific issue or behaviour of ComBat
• Implement metrics to quantify confound to target relationships
• Add tutorial notebooks, with few example use-cases
• Implement metrics to quantify the level of confounding in a given a sample
• Add tests for Residualize() with non-linear models
• Add tests for DummyDeconfounding() method

Twitter summary

Python library to handle #confounds/covariates in #machinelearning and #neuroscience, contribute to a great #openscience cause! github.com/raamana/confounds Pradeep Reddy Raamana @raamana_

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

confounds

Please read and follow the OHBM Code of Conduct

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

[raamana]

Hi @djarecka , where do I send the pitch slide to?

[djarecka]

here!

[raamana]

would this work?

[raamana]

let me know how to create a Discord channel

[raamana]

link to the specific issues that we would be focusing on this hackathon https://github.com/raamana/confounds/issues/18

[jrasero]

Hi Team,

I leave here the comment I left yesterday in the issue pinged in the previous message.

Yesterday I made some progress. I implemented a few things. A DeconfEstimator class, which first deconfounds the data and then runs a passed estimator, a deconfounded_cv_predict function to get predictions in a cross-validation scheme including deconfounding, and deconfounded_cv_score, the same but yielding the performance scores.

I also added a few tests to these new funcionalities.

Please, take a look, and let me know if you see these OK. I can pull request all of this if you want.