Hybrid deep learning for mental workload classification using EEG with enhanced preprocessing and interpretability

Clinical Snapshot

Key Findings

• P Value: Not reported

• Effect Size: Average accuracy of 83.9% across subjects, reported as highest among baseline comparators

• Primary Outcome: Four-class mental workload classification accuracy under leave-one-subject-out cross-validation

• Nnt Or Sensitivity: Per-class sensitivity, specificity, and F1-scores not reported in abstract; baseline comparison margins not quantified

• Confidence Interval: Not reported

Clinical Application

Current feasibility for clinical deployment is low. EEG acquisition requires specialised equipment, trained operators, controlled environments, and significant preprocessing infrastructure. Real-time implementation latency is not assessed. The 10-second window requirement introduces inherent temporal lag. Wearable dry-electrode EEG systems may reduce barriers but introduce additional noise challenges not addressed here. No TGA-registered EEG-based workload monitoring system currently exists for clinical use in Australia. RACGP and AHPRA guidelines do not address cognitive workload monitoring technology. Potential relevance exists for Safe Work Australia frameworks in high-risk occupational settings and for Australian aviation safety regulators (CASA). Research translation would require NHMRC-funded prospective validation studies in Australian populations before any clinical or regulatory consideration. PBS listing is not applicable at this stage. Potentially applicable to operators in safety-critical environments (healthcare workers, aviation personnel, military) where real-time cognitive workload monitoring could inform task allocation or fatigue management; not yet applicable to patient populations

Abstract

Mental workload classification is critical in safety-sensitive fields such as healthcare and aviation. However, electroencephalography-based approaches still face challenges with generalizability, noise robustness, and interpretability. In this study, we propose an integrated hybrid deep learning framework to address these limitations and enable robust, interpretable electroencephalography-based mental workload classification. The proposed approach uses a Variational Autoencoder to enhance noise reduction and feature extraction from band-wise topographical videos, a Convolutional Block Attention Module to adaptively focus on important spatial-channel Electroencephalogram features, and a Bidirectional Long Short-Term Memory network to capture complex temporal dependencies under leave-one-subject-out cross-validation. We conducted ablation studies to identify each architecture component's contribution and sensitivity analyses to determine the optimal parameters. The model achieved the highest overall accuracy among the baselines and reached an average accuracy of 83.9% across subjects for classifying four mental workload levels. Ablation studies confirmed the added value of all three architecture components for improving performance. Sensitivity analyses identified optimal parameters, including a 10-second window length that balances temporal context and specificity. For better interpretability and neurophysiological insight, we used Gradient-weighted Class Activation Mapping to visualize key frontal-parietal brain regions and frequency bands associated with workload dynamics. Future research could explore adaptive windowing strategies, multimodal data integration, cross-dataset benchmarking, and evaluation against transformer-based and graph neural network architectures under consistent subject-independent evaluation settings to further enhance model generalizability.

References

1. 1.Abdelrahman, O., XinYing, C., Malik, E. F., Wah, K. K., Lin, C. Z., & Lin, T. W. (2026). Hybrid deep learning for mental workload classification using EEG with enhanced preprocessing and interpretability. PLoS ONE. https://doi.org/10.1371/journal.pone.0352882