Skip to content

GSoC Ideas

Jump to bottom Edit New page
Eric Larson edited this page Feb 4, 2022 · 41 revisions

MNE-Python

MNE-Python is planning to participate in the GSOC 2021 under the 🐍 Python Software Foundation (PSF) umbrella.

Note: If you are not currently pursuing research activities in MEG or EEG and do not use or do not plan to use MNE-Python for your own research, our GSoC might not be for you. Our projects require domain-specific interest and are not simple coding jobs.

About MNE-Python

MNE-Python is a pure Python package for preprocessing and analysis of M/EEG data. For more information, see our homepage.

Contact

For modes of communication see our getting help page. It's a good idea to introduce yourself on our Discourse forum to get in touch with potential mentors before submitting an application.

Getting Started

Writing your GSoC application

Additional reading

Project ideas

We list some potential project ideas below, but we welcome other ideas that could fit within the scope of the project!

1. Facilitate access to open EEG/MEG databases

Difficulty

Medium (requires knowledge of M/EEG data)

Possible mentors

Denis Engemann, Alex Gramfort, Eric Larson, Daniel McCloy

Goal

The aim of this project is to improve the access to open EEG/MEG databases via the mne.datasets module, in other words, improve our dataset fetchers. There is physionet, but much more. Having a consistent API to access multiple data source would be great.

Subgoals

See https://github.com/mne-tools/mne-python/issues/2852 and https://github.com/mne-tools/mne-python/issues/3585 for some ideas, or:

  • MMN dataset (http://www.fil.ion.ucl.ac.uk/spm/data/eeg_mmn/ ) used for tutorial/publications applying DCM for ERP analysis using SPM.
  • Human Connectome Project Datasets (http://www.humanconnectome.org/data/ ). Over a 3-year span (2012-2015), the Human Connectome Project (HCP) scanned 1,200 healthy adult subjects. The available data includes MR structural scans, behavioral data and (on a subset of the data) resting state and/or task MEG data.
  • Kymata Datasets (https://kymata-atlas.org/datasets). Current and archived EMEG measurement data, used to test hypotheses in the Kymata atlas. The participants are healthy human adults listening to the radio and/or watching films, and the data is comprised of (averaged) EEG and MEG sensor data and source current reconstructions.
  • http://www.brainsignals.de/ A website that lists a number of MEG datasets available for download.
  • BNCI Horizon (http://bnci-horizon-2020.eu/database/data-sets) has several BCI datasets

2. Source-time-frequency visualization of brain data

Difficulty

Medium (requires visualization and time-frequency analysis knowledge)

Possible mentors

Mainak Jas, Britta Westner, Sarang Dalal, Denis Engemann, Alex Gramfort, Eric Larson

Goal

Implement viewer for interactive visualization of volumetric source-time-frequency (5-D) maps on MRI slices (orthogonal 2D viewer). NutmegTrip (written by Sarang Dalal) provides similar functionality in Matlab in conjunction with FieldTrip.

Example of NutmegTrip's source-time-frequency mode in action (click for link to YouTube): NutmegTrip source-time-frequency viewer example

Subgoals

  • extend source time series (4-D) viewer created by Mainak Jas to a time-frequency (5-D) viewer
  • Allow plotting surface data in volume using cortical ribbon (see closed glass brain PR: https://github.com/mne-tools/mne-python/pull/4496)
  • capability to output source map animations directly as GIFs
  • ability to plot on 3D rendered brains not necessary at this stage, but would be helpful if it is written such that it can be extended in the future
  • extra credit: plot corresponding EEG/MEG topography at selected time point. (Not in video, but this can be done in NutmegTrip as well – very useful as a sanity check for source results.)

3. Facilitate cloud computing

Possible mentors

Alex Gramfort, Eric Larson

Difficulty

Hard (requires cloud computing and Python knowledge)

Goal

Currently, cloud computing with M/EEG data requires multiple manual steps, including remote environment setup, data transfer, monitoring of remote jobs, and retrieval of output data/results. These steps are usually not specific to the analysis of interest, and thus should be something that can be taken care of by MNE.

Subgoals

  • Leverage dask and joblib or other libs to allow simple integration with MNE processing steps. Ideally this would be achieved in practice by:
    • One-time (or per-project) setup steps, setting up host keys, access tokens, etc.
    • In code, switch to cloud computing rather than local computing via a simple change of n_jobs parameter, and/or context manager like with use_dask(...): ....
  • Develop a (short as possible) example that shows people how to run a minimal task remotely, including setting up access, cluster, nodes, etc.
  • Adapt MNE-biomag-group-demo code to use cloud computing (optionally, based on config) rather than local resources.

4. Integrate OpenMEEG via improved Python bindings

Difficulty

Medium (requires working with C++ bindings)

Possible mentors

Alex Gramfort, Eric Larson

Goal

OpenMEEG is a state-of-the art solver for forward modeling in the field of brain imaging with MEG/EEG. It solves numerically partial differential equations (PDE). It is written in C++ with Python bindings written in SWIG. The ambition of the project is to integrate OpenMEEG into MNE offering to MNE the ability to solve more forward problems (cortical mapping, intracranial recordings, etc.).

Subgoals

  • Revamp Python bindings (remove useless functions, check memory managements, etc.)
  • Write example scripts for OpenMEEG that automatically generate web pages as for MNE
  • Package OpenMEEG for Debian/Ubuntu
  • Update the continuous integration system
Add a custom footer
Add a custom sidebar
Clone this wiki locally