Machine Learning Based Assessment of Mental Stress Using Wearable Sensors
Journal on Electronic and Automation Engineering · 01.05.2026 · 5 (2) · 136
DOI: 10.46632/jeae/5/2/18 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal on Electronic and Automation Engineering, online erschienen am 01.05.2026. Verfasst von Usha Shree A. Die Arbeit nennt 37 Literaturangaben.
Abstract (Originalsprache)
Mental stress has become a significant public health concern, contributing to hypertension, cardiovascular disease, and psychological disorders. This project presents a wearable, real-time stress monitoring system that integrates physiological sensing with machine learning (ML) to classify stress levels as Low, Moderate, or High. The system is built around an Arduino Uno microcontroller and employs multiple sensors: an SpO2/Heart Rate sensor (MAX30102) for blood oxygen saturation and Heart Rate Variability (HRV), a DHT11 for ambient temperature and humidity, and a NEO-6M GPS module for location-based stress mapping. Data is displayed on an LCD and transmitted to the cloud via an ESP8266 Wi-Fi module for ML-based analysis. Using a Multi-Layer Perceptron (MLP) neural network, the system achieved a classification accuracy of 95.8%. An alert mechanism (buzzer + cloud notification) triggers on detection of elevated stress. The project demonstrates a low-cost, scalable IoT-ML solution for preventative healthcare and real-time stress management.
Zitieren
APA
A, U. S. (2026). Machine Learning Based Assessment of Mental Stress Using Wearable Sensors. Journal on Electronic and Automation Engineering, 5(2), 136. https://doi.org/10.46632/jeae/5/2/18
Vancouver
A US. Machine Learning Based Assessment of Mental Stress Using Wearable Sensors. Journal on Electronic and Automation Engineering. 2026;5(2):136. doi:10.46632/jeae/5/2/18
BibTeX
@article{A2026,
author = {A, Usha Shree},
title = {Machine Learning Based Assessment of Mental Stress Using Wearable Sensors},
journal = {Journal on Electronic and Automation Engineering},
year = {2026},
volume = {5},
number = {2},
pages = {136},
doi = {10.46632/jeae/5/2/18}
}
EUVIMED