Discover datasets

STRUM is a dyadic neuroergonomics dataset in which pairs of participants perform a collaborative urban-reconnaissance task while a time-synchronized multimodal sensor suite records both members simultaneously via Lab Streaming Layer. It covers 33 sessions and 66 participants, with high-density EEG, EMG, ECG/EOG, motion capture, eye-tracking, force plate and controller behavior. Each participant has their own subject directory, and the pairing is encoded in participants.tsv so that analyses can split by pair.

Participants

66

Channels

207 (biosemi)

HED

v8.4.0

Size

512 GB

Version

v1.0.1

Updated

Sep 28, 2026

A NEMAR re-host and EMG-BIDS conversion of the public Meta emg2pose benchmark, pairing 2 kHz, 16-channel wrist surface electromyography with hand pose labels from a 26-camera motion-capture system. The release spans 193 users, about 370 hours, 29 behavioral stages and roughly 80 million labeled frames. It supports research on estimating hand pose from surface EMG.

Participants

193

Size

740 GB

Version

v1.0.3

Updated

Aug 31, 2026

SeizeIT2 is a multicenter, prospective dataset of wearable behind-the-ear EEG, ECG, EMG and movement recordings from 125 adult and pediatric patients with focal epilepsy. Patients were monitored in five European epilepsy monitoring units with the Sensor Dot wearable device alongside standard video-EEG. The dataset contains about 11,640 hours of wearable data with 886 recorded focal seizures, annotated in BIDS-score compatible format, and supports the development and validation of seizure detection algorithms.

Participants

125

Channels

2

Citations

4

Size

117 GB

Version

v1.0.0

Updated

Jun 28, 2026

High-density surface EMG (HDsEMG) recordings from the right tibialis anterior of a single healthy participant performing submaximal isometric ankle dorsiflexions at 10 to 70 percent of maximal voluntary contraction. Signals were acquired with two 64-electrode arrays (128 electrodes total) alongside force measurements. The dataset includes labeled motor unit spike trains derived from concurrent invasive EMG, and supports motor unit decomposition research.

Participants

1

Size

1.34 GB

Version

v1.0.1

Updated

Jun 1, 2026

A BIDS-formatted high-density surface EMG (HDsEMG) dataset from 16 young volunteers performing submaximal (10-80% MVC) isometric ankle dorsiflexions (tibialis anterior, n=8) or knee extensions (vastus lateralis, n=8). Signals were recorded with four 64-electrode arrays (256 channels in total) together with force signals, supporting the decomposition of extensive motor unit samples and the study of rate coding. It is a BIDS conversion of the data from Avrillon et al. 2024, produced with the MUniverse software.

Participants

16

Size

5.51 GB

Version

v1.0.1

Updated

Jun 1, 2026

A BIDS-formatted dataset of high-density surface EMG (HDsEMG) recordings from the right tibialis anterior of six healthy male participants. Four 64-channel grids (256 electrodes) recorded trapezoidal isometric dorsiflexion contractions at 30% and 50% of maximal voluntary contraction, together with force signals. The dataset supports studies of electrode grid design and motor unit identification.

Participants

6

Size

448 MB

Version

v1.0.1

Updated

Jun 1, 2026

NEMAR Dataset nm000107: wrist - Wrist movement control from EMG

Participants

100

Size

24.9 GB

Version

v2.0.0

Updated

Feb 23, 2026

NEMAR Dataset nm000106: handwriting - Handwriting movement detection from EMG

Participants

100

Size

45.3 GB

Version

v2.0.0

Updated

Feb 23, 2026

NEMAR Dataset nm000105: discrete_gestures - Discrete hand gesture detection from EMG

Participants

100

Size

20.6 GB

Version

v2.0.0

Updated

Feb 23, 2026

NEMAR Dataset nm000104: emg2qwerty - EMG-based typing detection

Participants

108

Citations

1

Size

223 GB

Version

v2.0.0

Updated

Feb 23, 2026