# PhysioMio: bilateral and longitudinal HD-sEMG dataset of 16 hand gestures from 48 stroke patients

## TL;DR

This paper introduces a dataset of high-density EMG recordings from stroke patients performing hand gestures, capturing muscle activity during rehabilitation.

## Contribution

The dataset is the first to provide longitudinal HD-sEMG data from both healthy and impaired arms of stroke patients.

## Key findings

- The dataset includes recordings from 48 stroke patients performing 16 hand gestures.
- Longitudinal data captures muscle activation patterns at different rehabilitation stages.
- The dataset supports research on neuromuscular deficits and recovery in stroke patients.

## Abstract

The PhysioMio dataset presented in this paper provides longitudinal high-density surface electromyography (HD-sEMG) recordings from both the healthy and impaired forearms of 48 stroke patients with arm paresis, captured during the performance of 16 distinct hand gestures. Patients were recorded at regular intervals during their individual inpatient rehabilitation stay, resulting in an average of three recording sessions at different stages of post-stroke rehabilitation per patient. The HD-sEMG signals were collected using a dry 64-electrode array positioned around the forearm, enabling observation of muscle activation patterns. The healthy arm was recorded during the first week to serve as a reference, while subsequent recordings focused solely on the impaired arm. This data can offer insights into neuromuscular deficits related to stroke and allow for comparative analysis between healthy and impaired arms. This dataset serves as a valuable resource for studying motor impairment and recovery potential in stroke-induced arm paresis, supporting advancements in personalized rehabilitation and assistive technologies.