Wearable Electrodermal Activity Sensor for Real-Time Stress Detection Using Machine Learning

Salvador Santos, Joana Sousa, João Ferreira

2025

Abstract

This paper discusses the design and implementation of a wearable electrodermal activity (EDA) sensor intended to detect subtle changes in skin conductivity, which are indicative of emotional states such as stress and anxiety, thus monitoring stress and arousal levels through advanced machine learning techniques. The device incorporates innovative conductive lycra combined with silver-silver chloride (Ag/AgCl) electrodes, enabling optimal skin contact and enhancing signal reliability. This integration allows for effective measurement of EDA. Utilizing the XGBoost algorithm, our machine learning model was trained on the ASCERTAIN dataset, achieving an overall accuracy of approximately 77% in predicting levels of arousal. While the model exhibited some challenges in predicting intermediate arousal states, it demonstrated strong precision and recall for extreme levels of arousal, underscoring its potential applications in mental health monitoring and human-computer interaction. The capabilities of this wearable technology for continuous and long-term health monitoring pave the way for further research into stress assessment and the understanding of emotional responses, emphasizing its relevance in enhancing psychological well-being.

Paper Citation

in Harvard Style

Santos S., Sousa J. and Ferreira J. (2025). Wearable Electrodermal Activity Sensor for Real-Time Stress Detection Using Machine Learning. In Proceedings of the 18th International Joint Conference on Biomedical Engineering Systems and Technologies - Volume 1: BIODEVICES; ISBN 978-989-758-731-3, SciTePress, pages 188-196. DOI: 10.5220/0013257900003911

in Bibtex Style

@conference{biodevices25,
author={Salvador Santos and Joana Sousa and João Ferreira},
title={Wearable Electrodermal Activity Sensor for Real-Time Stress Detection Using Machine Learning},
booktitle={Proceedings of the 18th International Joint Conference on Biomedical Engineering Systems and Technologies - Volume 1: BIODEVICES},
year={2025},
pages={188-196},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013257900003911},
isbn={978-989-758-731-3},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 18th International Joint Conference on Biomedical Engineering Systems and Technologies - Volume 1: BIODEVICES
TI - Wearable Electrodermal Activity Sensor for Real-Time Stress Detection Using Machine Learning
SN - 978-989-758-731-3
AU - Santos S.
AU - Sousa J.
AU - Ferreira J.
PY - 2025
SP - 188
EP - 196
DO - 10.5220/0013257900003911
PB - SciTePress