A Consecutive Multi-Day High-Density Surface Electromyography Dataset Comprising 7 Grasps and 11 Gestures

Abstract Surface electromyography (sEMG) records muscle electrical signals and reflects neuromuscular physiological behaviors. Recently, high-density sEMG (HD-sEMG), which allows non-invasive identification of motor unit action potential trains (MUAPTs) and direct access to underlining neural drive...

Full description

Saved in:
Bibliographic Details
Main Authors: Shutian Yang, Chen Chen, Dongxuan Li, Xiangyang Zhu
Format: Article
Language:English
Published: Nature Portfolio 2025-08-01
Series:Scientific Data
Online Access:https://doi.org/10.1038/s41597-025-05733-y
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Abstract Surface electromyography (sEMG) records muscle electrical signals and reflects neuromuscular physiological behaviors. Recently, high-density sEMG (HD-sEMG), which allows non-invasive identification of motor unit action potential trains (MUAPTs) and direct access to underlining neural drive derived from the spinal cord, becomes a research hotspot. However, datasets comprising HD-sEMG signals remain limited, especially for multi-day conditions, leading to the lack of long-term investigation of motor neuron activities. This paper presents a 320-channel HD-sEMG dataset, CEMHSEY (ConsecutivE Multi-day High-density Surface ElectromyographY), recorded from forearm muscles and across 11 consecutive days. The dataset consists of two sub-datasets as: an isometric contraction dataset containing 13 subjects performing 7 grasps under 3 different contraction force levels (named GRASP) and a hand gesture dataset with 6 subjects performing 11 hand gestures (named GESTURE). The dataset was validated with the usability of force regression, hand gesture recognition, and motor unit decoding. In addition, the multi-day data provide support for developing robust human-machine interfaces as well as analyzing neuromuscular modulation.
ISSN:2052-4463