Task-related EEG and HRV entropy factors under different real-world fatigue scenarios

NEUROCOMPUTING

卷:311

Article

Lin, Chin-Teng¹  Nascimben, Mauro²  King, Jung-Tai²  Wang, Yu-Kai¹ 

[1] Univ Technol Sydney, FEIT, Ctr Artificial Intelligence, CIBCI Lab, Sydney, NSW, Australia

[2] Natl Chiao Tung Univ, Brain Res Ctr, Hsinchu, Taiwan

关键词: Human performance;    Entropy analysis;    Alertness prediction;    EEG;    HRV;    Psychomotor vigilance task;   

DOI  :  10.1016/j.neucom.2018.05.043

来源: Elsevier

【 摘 要 】

We classified the alertness levels of 17 subjects in different experimental sessions in a six-month longitudinal study based on a daily sampling system and related alertness to performance on a psychomotor vigilance task (PVT). As to our best knowledge, this is the first EEG-based longitudinal study for real-world fatigue. Alertness and PVT performance showed a monotonically increasing relationship. Moreover, we identified two measures in the entropy domain from electroencephalography (EEG) and heart rate variability (HRV) signals that were able to identify the extreme classes of PVT performers. Wiener entropy on selected leads from the frontal-parietal axis was able to discriminate the group of best performers. Sample entropy from the HRV signal was able to identify the worst performers. This joint EEG-HRV quantification provides complementary indexes to indicate more reliable human performance. (C) 2018 Elsevier B.V. All rights reserved.

【 授权许可】

Free