EEGDash: An open-source platform for machine learning on public neurophysiological data

Abstract

Public recordings of electroencephalography (EEG), magnetoencephalography (MEG), intracranial EEG (iEEG), functional near-infrared spectroscopy (fNIRS) and electromyography (EMG) are easy to find and hard to train on. Turning a deposit into model input still means writing download, parsing, repair and windowing code for each dataset, and a dataset that meets a metadata standard can still fail to load. EEGDash is an open-source registry over public Brain Imaging Data Structure (BIDS) datasets from the OpenNeuro and NEMAR archives and from local deposits. Each dataset becomes a Python class; format defects are repaired lazily; signal handling, windowing and format checking are delegated to MNE-Python, Braindecode and the official BIDS validator. Semantic search, a web browser, publication-derived dataset tags and a feature-extraction module share the registry. EEGDash indexes the metadata of 892 public BIDS datasets (43,911 participants and 126,672 hours across the five modalities), loads any of them into a PyTorch dataset without custom data-loading code, and shows on 503 OpenNeuro EEG datasets that passing the BIDS validator does not predict whether a dataset loads (Spearman ρ= −0.05), so an archive built for machine learning has to track both.

An automated ranking platform for machine learning regression models for meat spoilage prediction using multi-spectral imaging and metabolic profiling
Lucia Estelles-Lopez, Athina Ropodi, Dimitris Pavlidis +6
Food Research International9 Oct 2026Peer-reviewedLicence not stated86 citedGeneral biology StatisticsHAL
Time-split validation and cross-platform transferability of an explainable MALDI-TOF MS machine-learning model for carbapenem-resistant Klebsiella pneumoniae screening
Kun Li, Xue Yang, Long Li +1
Frontiers in Microbiology28 Oct 2026Peer-reviewedCC BY 4.0Full textMicrobiologyCrossref
Subject-independent classification of ADHD from pediatric EEG: a leakage-aware deep learning framework with interpretable neurophysiological mapping
M. Geethapriya, Kalyanaraman Pattabiraman
Frontiers in Neuroscience6 Oct 2026Peer-reviewedCC BY 4.0Full textNeuroscienceCrossrefEurope PMC
Stimulatrix: An open-source automated platform for high-throughput functional characterization of engineered contractile tissues
M. Winkelbauer, J. Bast, E. S. Casalino +10
bioRxiv24 Sep 2026PreprintCC BYFull textSystems & synthetic biologybioRxiv
Open-source virtual reality platform for head-fixed adult zebrafish
A. Javier, S. Narayanan, L. Anneser +2
bioRxiv19 Oct 2026PreprintCC BYFull textNeurosciencebioRxiv

Cite

BibTeX RIS

BibTeX

@misc{aristimunha2026eegdash,
  title = {{EEGDash: An open-source platform for machine learning on public neurophysiological data}},
  author = {Bruno Aristimunha and Aviv Dotan and Pierre Guetschel and Aman Jaiswal and Gal Ashkenazi and Dung Truong and Kuntal Kokate and Amit Majumdar and Oren Shriki and Arnaud Delorme},
  howpublished = {arXiv},
  year = {2026},
  month = oct,
  doi = {10.48550/arxiv.2606.16041},
  eprint = {2606.16041},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2606.16041},
  note = {CC BY 4.0},
}

RIS

TY  - UNPB
TI  - EEGDash: An open-source platform for machine learning on public neurophysiological data
AU  - Bruno Aristimunha
AU  - Aviv Dotan
AU  - Pierre Guetschel
AU  - Aman Jaiswal
AU  - Gal Ashkenazi
AU  - Dung Truong
AU  - Kuntal Kokate
AU  - Amit Majumdar
AU  - Oren Shriki
AU  - Arnaud Delorme
PY  - 2026
DA  - 2026/10/11
PB  - arXiv
DO  - 10.48550/arxiv.2606.16041
UR  - https://arxiv.org/abs/2606.16041
AB  - Public recordings of electroencephalography (EEG), magnetoencephalography (MEG), intracranial EEG (iEEG), functional near-infrared spectroscopy (fNIRS) and electromyography (EMG) are easy to find and hard to train on. Turning a deposit into model input still means writing download, parsing, repair and windowing code for each dataset, and a dataset that meets a metadata standard can still fail to load. EEGDash is an open-source registry over public Brain Imaging Data Structure (BIDS) datasets from the OpenNeuro and NEMAR archives and from local deposits. Each dataset becomes a Python class; format defects are repaired lazily; signal handling, windowing and format checking are delegated to MNE-Python, Braindecode and the official BIDS validator. Semantic search, a web browser, publication-derived dataset tags and a feature-extraction module share the registry. EEGDash indexes the metadata of 892 public BIDS datasets (43,911 participants and 126,672 hours across the five modalities), loads any of them into a PyTorch dataset without custom data-loading code, and shows on 503 OpenNeuro EEG datasets that passing the BIDS validator does not predict whether a dataset loads (Spearman ρ= −0.05), so an archive built for machine learning has to track both.
ER  -

Abstract

Related

Cite