EEG and ECG Wearable Biosensor-Based Affective State Analysis Using Deep Learning over 6G IoT Healthcare Networks
Biosensors · 23.07.2026 · 16 (8) · 400
DOI: 10.3390/bios16080400 ↗Kurz gefasst
Fachartikel in der Zeitschrift Biosensors, online erschienen am 23.07.2026. Verfasst von Hazal Su Bıçakcı Yeşilkaya und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Electrical and Electronics Engineering, Faculty of Engineering and Architecture, İzmir Bakırçay University, 35665 İzmir, Türkiye. Hauptthemen laut MEDLINE: Wearable Electronic Devices, Deep Learning, Electrocardiography, Electroencephalography. Die Arbeit nennt 93 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Physiological signal analysis using wearable biosensors like an electroencephalogram (EEG) and an electrocardiogram (ECG) is widely investigated for affective computing; however, the integration of deep learning-based affective computing within 6G-driven IoT healthcare infrastructures remains limited, with data transmission latency posing a significant challenge. This study proposes a framework for EEG and ECG-based affective state analysis over 6G IoT networks. We utilize an attention-based deep learning model for three-class emotion recognition from EEG signals, and a ResNet50-based convolutional neural network for three-class stress/affective state classification using ECG data. The framework’s communication performance is evaluated through ray tracing simulations in a virtual hospital environment at 7 GHz and 92 GHz bands. Experimental results on SEED and WESAD datasets demonstrate that the EEG model achieved 90.93% classification accuracy, while the ECG model yielded 82.26% validation and 65.64% test accuracy. Wireless analysis showed RMS delay spread values of 6.53 ns (7 GHz) and 3.32 ns (92 GHz), with correlation bandwidths of 30.63 MHz and 60.24 MHz, respectively. These findings demonstrate the feasibility of integrating wearable biosensor-based affective state analysis with 6G-oriented IoT healthcare communication frameworks, providing a robust foundation for future personalized health and human state monitoring applications.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Electrical and Electronics Engineering, Faculty of Engineering and Architecture, İzmir Bakırçay University, 35665 İzmir, Türkiye.
- Department of Computer Engineering, Faculty of Engineering and Architecture, İzmir Bakırçay University, 35665 İzmir, Türkiye.
- Department of Biomedical Engineering, Faculty of Engineering and Architecture, İzmir Katip Çelebi University, Balatcik Campus, 35620 İzmir, Türkiye.
ORCID-Profile: Hazal Su Bıçakcı Yeşilkaya, Mete Özbaltan, Nihan Özbaltan, Cihat Şeker, Bartu Yeşilkaya, Bengisu Yalçınkaya
Zitieren
APA
Bıçakcı Yeşilkaya, H. S., Özbaltan, M., Özbaltan, N., Şeker, C., Yeşilkaya, B., & Yalçınkaya, B. (2026). EEG and ECG Wearable Biosensor-Based Affective State Analysis Using Deep Learning over 6G IoT Healthcare Networks. Biosensors, 16(8), 400. https://doi.org/10.3390/bios16080400
Vancouver
Bıçakcı Yeşilkaya HS, Özbaltan M, Özbaltan N, Şeker C, Yeşilkaya B, Yalçınkaya B. EEG and ECG Wearable Biosensor-Based Affective State Analysis Using Deep Learning over 6G IoT Healthcare Networks. Biosensors. 2026;16(8):400. doi:10.3390/bios16080400
BibTeX
@article{BakcYeilkaya2026,
author = {Bıçakcı Yeşilkaya, Hazal Su and Özbaltan, Mete and Özbaltan, Nihan and Şeker, Cihat and Yeşilkaya, Bartu and Yalçınkaya, Bengisu},
title = {EEG and ECG Wearable Biosensor-Based Affective State Analysis Using Deep Learning over 6G IoT Healthcare Networks},
journal = {Biosensors},
year = {2026},
volume = {16},
number = {8},
pages = {400},
doi = {10.3390/bios16080400}
}
EUVIMED