WhoBPyT - Whole-Brain Modelling in PyTorch

[JohnGriffiths]

Title

WhoBPyT - Whole-Brain Modelling in PyTorch

Short description and the goals for the OHBM BrainHack

WhoBPyT is a Python library for connectome-based neurophysiological modelling of neuroimaging and other macroscale brain data. At its core is a deep learning-based approach for model fitting and parameter estimation, that capitalizes on the in-built algorithmic differentiation and optimization functionality of PyTorch.

The current WhoBPyT code base is a consolidation of several projects that have used this computational technique to study mechanisms underlying brain dynamics in TMS-EEG, stereotactic EEG, resting-state fMRI.

We have completed a major refactor of the code base that generalizes several key functionalities and improves overall readability. The goal of this OHBM BrainHack project is to improve the first-time user experience with WhoBPyT, principally through docs and tutorial examples on computational modeling of EEG and fMRI data.

Link to the Project

https://github.com/GriffithsLab/whobpyt

Image/Logo for the OHBM brainhack website

https://drive.google.com/uc?export=view&id=1wJaFEEk6hDfeOCjix2v_BhmWmNak9VzO

Project lead

John Griffiths

Main Hub

Seoul

Link to the Project pitch

Pitch video will be added ahead of the event.

Other hubs covered by the leaders

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

Skills

Python
PyTorch
Github
EEG
fMRI
MNE
Nilearn
Numerical simulation
Reading+writing

Recommended tutorials for new contributors

OHBM brain modelling edu course - https://griffithslab.github.io/OHBM2024-educational-course/

Good first issues

• Test the EEG/fMRI modelling examples
• Run examples with alternative data
• Improve data visualization in examples
• Add any missing docstrings
• Add missing items to the README
• ...

Twitter summary

No response

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

PyTorchBrainModelling

Please read and follow the OHBM Code of Conduct

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

[sina-mansour]

@JohnGriffiths Many thanks for your project submission to OHBM brainhack 2024, the project looks good to go!

We look forward to seeing you in Seoul!