EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

TEANet: A Transpose-Enhanced Autoencoder Network for Stress Monitoring on Resource-Constrained Wearable Device

Md Santo Ali, Sapnil Sarker Bipro, Mohammod Abdul Motin, Sumaiya Kabir, Manish Sharma, Muhammad E. H. Chowdhury

IEEE Access · 2026 · 14 · 9560-9571

DOI: 10.1109/access.2025.3644400 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Access, 2026. Verfasst von Md Santo Ali und 5 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 42 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Qatar National Library (QNL), for Open Access Publication.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Mental stress poses a significant public health concern due to its detrimental effects on physical and mental well-being, necessitating the development of continuous stress monitoring tools for wearable devices. Blood volume pulse (BVP) signals, readily available in many smartwatches, offer a convenient and cost-effective solution for stress monitoring. This study presents a deep learning model, a Transpose-Enhanced Autoencoder Network (TEANet), for stress detection using BVP signals on resource-constrained devices. The proposed TEANet model was trained and validated utilizing a self-developed RUET SPML dataset, and the publicly available wearable stress and affect detection (WESAD) dataset. It achieves accuracies of 92.94% and 96.94%, F1 scores of 95.16% and 95.95%, and Cohen’s kappa values of 0.8181 and 0.9350 for RUET SPML, and WESAD datasets, respectively. The proposed model outperforms most of the existing models in the literature. Additionally, its performance is consistently superior when compared to state-of-the-art convolutional and transformer-based architectures. Additionally, integrated gradient attribution based explainability is explored to further explain the model’s transparency in decision-making across different frequency bands. The proposed TEANet effectively detects mental stress through BVP signals with high accuracy, making it a promising tool for continuous stress monitoring. Furthermore, deploying the proposed model on the Raspberry Pi 3B+ enhances its potential for reliable real-time stress monitoring using resource-constrained devices.

Förderung

  • Qatar National Library (QNL), for Open Access Publication

Zitieren

APA

Ali, M. S., Bipro, S. S., Motin, M. A., Kabir, S., Sharma, M., & Chowdhury, M. E. H. (2026). TEANet: A Transpose-Enhanced Autoencoder Network for Stress Monitoring on Resource-Constrained Wearable Device. IEEE Access, 14, 9560-9571. https://doi.org/10.1109/access.2025.3644400

Vancouver

Ali MS, Bipro SS, Motin MA, Kabir S, Sharma M, Chowdhury MEH. TEANet: A Transpose-Enhanced Autoencoder Network for Stress Monitoring on Resource-Constrained Wearable Device. IEEE Access. 2026;14:9560-9571. doi:10.1109/access.2025.3644400

BibTeX
@article{Ali2026,
  author = {Ali, Md Santo and Bipro, Sapnil Sarker and Motin, Mohammod Abdul and Kabir, Sumaiya and Sharma, Manish and Chowdhury, Muhammad E. H.},
  title = {TEANet: A Transpose-Enhanced Autoencoder Network for Stress Monitoring on Resource-Constrained Wearable Device},
  journal = {IEEE Access},
  year = {2026},
  volume = {14},
  pages = {9560--9571},
  doi = {10.1109/access.2025.3644400}
}
Kostenlos starten →