Towards Real-Time Cognitive State Monitoring Using an Explainable Data-Driven Deep Learning Framework on Physiological Signals
International Journal of Advanced Computer Science and Applications · 2026 · 17 (10)
DOI: 10.14569/ijacsa.2026.0171041 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Advanced Computer Science and Applications, 2026. Verfasst von Mohammed Ibrahim Altwijri.
Abstract (Originalsprache)
Cognitive states represent dynamic mental conditions that impact how individuals identify their environment, process information, and make decisions. These states are influenced by factors like stress levels, cognitive workload, and external stimuli, which can drastically affect task performance and safety. Therefore, precise examination of cognitive workload is critical in applications where aligning task requirements with human processing ability is vital. Over the past several years, deep learning models have excelled in cognitive state monitoring (e.g., fatigue, workload, attention, emotion) by automatically capturing complex features from physiological signals such as electroencephalography and eye-tracking, outperforming standard techniques for better human-computer interaction (HCI), personalized learning, and mental health insights, though challenges remain in data quality, labeling, and real-world generalizability. Moreover, explainable artificial intelligence techniques are incorporated widely to ensure robust and transparent model interpretability. In this study, we propose a Real-Time Cognitive State Monitoring Framework using Explainable Deep Learning (RTCSM-XDL) based on physiological signals. The proposed RTCSM-XDL model aims to achieve accurate and transparent cognitive state classification by integrating advanced data-driven learning and explainable artificial intelligence techniques. The data preprocessing stage includes data normalization, missing data handling, and categorical encoding to ensure high-quality and consistent inputs. To recognize the most useful physiological features, a hybrid feature selection strategy combining minimum redundancy maximum relevance, recursive feature elimination, and Boruta is employed. For cognitive state classification, a temporal residual gated unit with attention is designed to efficiently learn temporal dependencies and enable real-time inference. To address the interpretability complexity associated with deep learning methods, the proposed model utilizes LIME to provide human-understandable insights into model predictions. The experimental result analysis of the RTCSM-XDL technique takes place on a benchmark dataset, and the comparative results demonstrate improved performance over the existing approaches in terms of diverse metrics.
Zitieren
APA
Altwijri, M. I. (2026). Towards Real-Time Cognitive State Monitoring Using an Explainable Data-Driven Deep Learning Framework on Physiological Signals. International Journal of Advanced Computer Science and Applications, 17(10). https://doi.org/10.14569/ijacsa.2026.0171041
Vancouver
Altwijri MI. Towards Real-Time Cognitive State Monitoring Using an Explainable Data-Driven Deep Learning Framework on Physiological Signals. International Journal of Advanced Computer Science and Applications. 2026;17(10). doi:10.14569/ijacsa.2026.0171041
BibTeX
@article{Altwijri2026,
author = {Altwijri, Mohammed Ibrahim},
title = {Towards Real-Time Cognitive State Monitoring Using an Explainable Data-Driven Deep Learning Framework on Physiological Signals},
journal = {International Journal of Advanced Computer Science and Applications},
year = {2026},
volume = {17},
number = {10},
doi = {10.14569/ijacsa.2026.0171041}
}
EUVIMED