Paperdoi:10.1038/s41598-026-80282-x
A hybrid neural deep belief networks-gated recurrent unit algorithm with a leakage-aware evaluation protocol for ECG-HRV-based cognitive load classification
Scientific Reports, 2026
TL;DR
DBN–GRU HRV cognitive-load classifier on WESAD/SWELL with participant-disjoint evaluation: LOSO 68.4% accuracy / 0.614 macro-F1 vs a 52.8% / 0.230 majority baseline. Participant-relative features helped 13 of 15 people; architecture choice did not matter and zero-shot cross-dataset transfer fell below the majority baseline.
Abstract
Cognitive underload and overload can impair knowledge workers’ performance. Electrocardiogram-derived heart rate variability (HRV) offers a practical signal for assessing cognitive demand. However, sample-wise evaluation may reward participant recognition rather than cognitive-state classification. This study proposes and evaluates a hybrid deep belief network–gated recurrent unit (DBN–GRU) algorithm on WESAD and SWELL. The model combines restricted Boltzmann machine pre-training with a gated classifier head. The primary WESAD evaluations use participant-disjoint protocols and validation-based model selection. Under leave-one-subject-out validation, the model achieves 68.4% accuracy, 62.8% balanced accuracy and 0.614 macro-F1. A majority-class predictor achieves 52.8% accuracy and 0.230 macro-F1. Participant-relative features improve performance for 13 of 15 participants ( p = 0.0015), whereas adversarial suppression of participant identity does not improve classification. Participant-relative features require unlabelled calibration data from each new participant. None of the eight alternative architectures differs significantly from the proposed model. Zero-shot cross-dataset transfer falls below the target majority-class baseline in both directions. Fine-tuning underperforms direct training on the target dataset. These results indicate a larger effect of participant-relative features than of architectural choice under the evaluated conditions as a quality assurance metrics.
Verbatim from OpenAlex (opens in a new tab) · Abstract via OpenAlex (publisher-deposited); © the authors/publisher
Results on ChatPPG leaderboards
All leaderboardsNo results from this paper on ChatPPG leaderboards yet. Browse the leaderboards.
Tasks & signals
Tasks
Signals
Devices
Keywords
Datasets used
How it works
Cite this paper
Keikhosrokiani, P., Wang, N., Shi, J., Liu, W., & Fadilah, S. I. (2026). A hybrid neural deep belief networks-gated recurrent unit algorithm with a leakage-aware evaluation protocol for ECG-HRV-based cognitive load classification. Scientific Reports, 2026. https://doi.org/10.1038/s41598-026-80282-x
Keikhosrokiani, Pantea, Ning Wang, Jianshu Shi, Wanhui Liu, and Suzi Iryanti Fadilah. 2026. "A hybrid neural deep belief networks-gated recurrent unit algorithm with a leakage-aware evaluation protocol for ECG-HRV-based cognitive load classification." Scientific Reports, 2026. https://doi.org/10.1038/s41598-026-80282-x.
Keikhosrokiani, P., Wang, N., Shi, J., Liu, W. and Fadilah, S. I. (2026) 'A hybrid neural deep belief networks-gated recurrent unit algorithm with a leakage-aware evaluation protocol for ECG-HRV-based cognitive load classification', Scientific Reports, 2026. doi: 10.1038/s41598-026-80282-x.
Keikhosrokiani, Pantea, et al. "A hybrid neural deep belief networks-gated recurrent unit algorithm with a leakage-aware evaluation protocol for ECG-HRV-based cognitive load classification." Scientific Reports, 2026, 2026. https://doi.org/10.1038/s41598-026-80282-x.
P. Keikhosrokiani, N. Wang, J. Shi, W. Liu, and S. I. Fadilah, "A hybrid neural deep belief networks-gated recurrent unit algorithm with a leakage-aware evaluation protocol for ECG-HRV-based cognitive load classification," Scientific Reports, 2026, 2026, doi: 10.1038/s41598-026-80282-x.
@article{keikhosrokiani2026hybrid,
title = {{A hybrid neural deep belief networks-gated recurrent unit algorithm with a leakage-aware evaluation protocol for ECG-HRV-based cognitive load classification}},
author = {Keikhosrokiani, Pantea and Wang, Ning and Shi, Jianshu and Liu, Wanhui and Fadilah, Suzi Iryanti},
year = {2026},
journal = {Scientific Reports, 2026},
doi = {10.1038/s41598-026-80282-x}
}Added to the ChatPPG research hub
