Investigating Driver Fatigue versus Alertness Using the Granger Causality Network

Sensors

Wanzeng Kong³  Weicheng Lin³  Fabio Babiloni²  Sanqing Hu³  Gianluca Borghini⁴  Felipe Jimenez¹ 

[1] College of Computer Science, Hangzhou Dianzi University, Hangzhou 310018, China;;Department of Molecular Medicine, University of Rome “Sapienza”, Rome 00185, Italy;College of Computer Science, Hangzhou Dianzi University, Hangzhou 310018, China; E-Mails:;Department of Physiology and Pharmacology, University of Rome “Sapienza”, Rome 00185, Italy; E-Mail:

关键词: driving fatigue;    eeg;    granger causality;    frequency domain;    brain effective network;   

DOI  :  10.3390/s150819181

来源: mdpi

【 摘 要 】

Driving fatigue has been identified as one of the main factors affecting drivers’ safety. The aim of this study was to analyze drivers’ different mental states, such as alertness and drowsiness, and find out a neurometric indicator able to detect drivers’ fatigue level in terms of brain networks. Twelve young, healthy subjects were recruited to take part in a driver fatigue experiment under different simulated driving conditions. The Electroencephalogram (EEG) signals of the subjects were recorded during the whole experiment and analyzed by using Granger-Causality-based brain effective networks. It was that the topology of the brain networks and the brain’s ability to integrate information changed when subjects shifted from the alert to the drowsy stage. In particular, there was a significant difference in terms of strength of Granger causality (GC) in the frequency domain and the properties of the brain effective network i.e., causal flow, global efficiency and characteristic path length between such conditions. Also, some changes were more significant over the frontal brain lobes for the alpha frequency band. These findings might be used to detect drivers’ fatigue levels, and as reference work for future studies.

【 授权许可】

CC BY   
© 2015 by the authors; licensee MDPI, Basel, Switzerland.