AEDAPT - A multi-platform virtual desktop environment with fully-preinstalled software for reproducible EEG/MEG/MRI data analysis

Last updated on Dec 9, 2021

Title

AEDAPT - A multi-platform virtual desktop environment with fully-preinstalled software for reproducible EEG/MEG/MRI data analysis

Leaders

Collaborators

Brainhack Global 2021 Event

Brainhack Australasia

Project Description

What are you doing, for whom, and why? AEDAPT provides a containerised data analysis graphical environment (based on Neurodesk; https://neurodesk.github.io/) to facilitate reproducible analysis of neuroimaging data by technical and less-technical users alike. It also provides interoperability with other data acquisition/storage/analysis/sharing platforms and utilities. One such utility developed by the AEDAPT team is SOVABIDS – a graphical converter from raw EEG data to the BIDS standard. AEDAPT and SOVABIDS are funded by the Australian Research Data Commons (ARDC), Swinburne University, University of Queensland and Sydney University. SOVABIDS was also funded by Google Summer of Code.

What makes your project special and exciting? Analysis pipelines for EEG/MEG/MRI data typically rely on specific versions of packages and software, and are dependent on their native operating system. These dependencies mean that a working analysis pipeline may fail or produce different results on a new computer, or even on the same computer after a software update. The frequent release of versions also means that researchers need to spend a lot of time on installing and upgrading software. AEDAPT provides a platform in which anyone, anywhere, using any computer can reproduce the original research findings given the original data and analysis code. Moreover, SOVABIDS allows the conversion of the original data to BIDS, thus making it easier for others to re-analyse the data using other tools.

AEDAPT and SOVABIDS are intended to serve researchers and developers in Australia and worldwide, and would be maintained by the community as open-source projects. It is a great opportunity to familiarise yourself with the neuroscience open-source community, as we are working closely with prominent developers and scientists all around the world.

How to get started? Head over to the resource links below and try Neurodesk (the technological platform on which AEDAPT runs) and SOVABIDS. Both are already tested and usable.

Where to find key resources?

https://neurodesk.github.io/ https://sovabids.readthedocs.io/

Goals for Brainhack Global

Good first issues

  1. Issue one: Download Neurodesk from https://neurodesk.github.io/ and test it for your EEG/MEG work. This is the technological platform of AEDAPT. Report any problems in: https://github.com/NeuroDesk/neurodesk.github.io/discussions

  2. Issue two: If you have EEG data in the Neuroscan’s (.cnt) format, download SOVABIDS from https://sovabids.readthedocs.io/en/latest/README.html#installation and test it. Alternatively, you can use publicly available EEG .cnt raw data (see below under “data to use”) Report any problems in: https://github.com/yjmantilla/sovabids/issues

  3. Issue three: Adding EEG/MEG workflow examples to our documentation page (workflows that use the tools preinstalled in AEDAPT). There is also the possibility to record video tutorials. https://neurodesk.github.io/tutorials/electrophysiology/ Possible idea: a workflow to convert raw EEG data in the .cnt format to BIDS standard using SOVABIDS, and then preprocess it using EEGLAB (available in AEDAPT).

Communication channels

https://mattermost.brainhack.org/brainhack/channels/virtual-neuro-machine-vnm

Skills

For good first issues:

For main goals:

Onboarding documentation

No response

What will participants learn?

BIDS (https://bids.neuroimaging.io/) Docker containers Github workflows Compiling MATLAB code JSON-RPC API

Data to use

Publicly available Neuroscan’s raw EEG data in .cnt format

Driving simulator experiment Paper: https://www.nature.com/articles/s41597-019-0027-4 Data: https://figshare.com/articles/dataset/The_original_EEG_data_for_driver_fatigue_detection/5202739

Motor imagery experiment Paper: https://www.nature.com/articles/s41597-020-0535-2 Data: https://doi.org/10.7910/DVN/RBN3XG

Number of collaborators

1

Credit to collaborators

Project contributors will be listed on the website, on Github, and potentially be part of publications originating from the project.

Image

AEDAPT_BrainHack 2021_2

Type

coding_methods, documentation, pipeline_development, tutorial_recording

Development status

2_releases_existing

Topic

connectome, data_visualisation, diffusion, EEG_EventRelatedResponseModelling, EEG_source_modelling, hypothesis_testing, ICA, neural_decoding, neural_encoding, neural_networks, PCA, physiology, reproducible_scientific_methods, statistical_modelling, systems_neuroscience

Tools

BIDS, Brainstorm, Datalad, FieldTrip, Freesurfer, Jupyter, MNE, Nipype, SPM

Programming language

containerization, documentation, Matlab, Python, shell_scripting, unix_command_line, html_css

Modalities

behavioral, ECG, ECOG, EEG, eye_tracking, MEG, MRI, TDCS, TMS

Git skills

2_branches_PRs

Anything else?

More information about AEDAPT can be found on our website: aedapt.net

Things to do after the project is submitted.


Date
Jan 1, 0001 12:00 AM