EEG classification of driver mental states by deep learning

Zeng, Hong; Chen, Yang; Dai, Guojun; Qin, Feiwei; Zhang, Jianhai; Kong, Wanzeng

doi:10.1007/s11571-018-9496-y

Cited by 169 publications

(102 citation statements)

References 50 publications

Supporting

Mentioning

101

Contrasting

Order By: Relevance

“…Recent conventional works with respect to the BCI-based detection of mental states have focused on accurate classification of user mental states using advanced machine learning algorithms and deep learning architecture [60][61][62][63][64][65]. Recognizing the mental conditions of drivers or pilots is a critical issue in systems using AI technology, such as autonomous vehicles and autopilot.…”

Section: Discussionmentioning

confidence: 99%

Classification of Drowsiness Levels Based on a Deep Spatio-Temporal Convolutional Bidirectional LSTM Network Using Electroencephalography Signals

Jeong

Lee

et al. 2019

Brain Sciences

View full text Add to dashboard Cite

Non-invasive brain-computer interfaces (BCI) have been developed for recognizing human mental states with high accuracy and for decoding various types of mental conditions. In particular, accurately decoding a pilot’s mental state is a critical issue as more than 70% of aviation accidents are caused by human factors, such as fatigue or drowsiness. In this study, we report the classification of not only two mental states (i.e., alert and drowsy states) but also five drowsiness levels from electroencephalogram (EEG) signals. To the best of our knowledge, this approach is the first to classify drowsiness levels in detail using only EEG signals. We acquired EEG data from ten pilots in a simulated night flight environment. For accurate detection, we proposed a deep spatio-temporal convolutional bidirectional long short-term memory network (DSTCLN) model. We evaluated the classification performance using Karolinska sleepiness scale (KSS) values for two mental states and five drowsiness levels. The grand-averaged classification accuracies were 0.87 (±0.01) and 0.69 (±0.02), respectively. Hence, we demonstrated the feasibility of classifying five drowsiness levels with high accuracy using deep learning.

show abstract

Section: Discussionmentioning

confidence: 99%

Classification of Drowsiness Levels Based on a Deep Spatio-Temporal Convolutional Bidirectional LSTM Network Using Electroencephalography Signals

Jeong

Lee

et al. 2019

Brain Sciences

View full text Add to dashboard Cite

show abstract

“…Electroencephalogram (EEG) is often used to monitor brain activity that reveals corresponding changes of the CNS. According to the literature, there are four main brain rhythms extracted from EEG signals, which closely relate to visual fatigue: delta, theta, alpha, and beta brain waves . Physiological signals such as skin temperature (SKT), oxygen saturation (SpO 2 ), and electrocardiogram (ECG) provide insight into ANS activities and can be considered the quantification of visual fatigue .…”

Section: Introductionmentioning

confidence: 99%

“…According to the literature, there are four main brain rhythms extracted from EEG signals, which closely relate to visual fatigue: delta, theta, alpha, and beta brain waves. [13][14][15] Physiological signals such as skin temperature (SKT), oxygen saturation (SpO 2 ), and electrocardiogram (ECG) provide insight into ANS activities and can be considered the quantification of visual fatigue. [16][17][18] Heart rate variability (HRV), a measure of the variation in time between each heartbeat, was extracted from ECG and testified as a good marker reflecting visual fatigue.…”

Section: Introductionmentioning

confidence: 99%

Assessment of visual fatigue under LED tunable white light with different blue components

et al. 2020

View full text Add to dashboard Cite

Light‐emitting diode (LED) light source has high intensity emission of blue components absent in the daylight spectra and regulates human physiology and behavior. The aim of this study was to explore the effects of LED tunable white light with different blue‐component intensities on visual fatigue based on human eye photoreceptors. The short (S)‐cone and melanopsin illuminance were about 212% and 82% higher for blue‐enriched white light than blue‐less white light, respectively. The photopic illuminance was same for these two lights. The results revealed that blue‐enriched LED tunable white light with higher illuminance of S cones had a significant effect on visual fatigue. Participants experienced more eye discomfort under blue‐enriched white light accompanied with decreased vision function and changes in the autonomic nervous system. Visual acuity and tear film stability declined, and heart rhythm changed more significantly under blue‐enriched white light than blue‐less white light. While memory performance did not decline with more severe visual fatigue, improved memory performance under blue‐enriched white light may be due to enhanced alertness or arousal associated with high melanopsin illuminance. Our results suggest that blue‐enriched white light with higher illuminance of S cones and melanopsin has beneficial effects on cognitive performance, but it can induce relatively more visual fatigue.

show abstract

“… 2018 ; Zeng et al. 2018 ). Later on, attention models became modular and more executive (Johnston and Heinz 1978 ; Posner and Snyder 1975 ): attention is considered as a flexible mental process, where voluntary control of attention requires more resources than automatic attentional discrimination.…”

Section: Introductionmentioning

confidence: 99%

A cognitive brain–computer interface monitoring sustained attentional variations during a continuous task

2019

View full text Add to dashboard Cite

We introduce a cognitive brain-computer interface based on a continuous performance task for the monitoring of variations of visual sustained attention, i.e. the self-directed maintenance of cognitive focus in non-arousing conditions while possibly ignoring distractors and avoiding mind wandering. We introduce a visual sustained attention continuous performance task with three levels of task difficulty. Pairwise discrimination of these task difficulties from electroencephalographic features was performed using a leave-one-subject-out cross validation approach. Features were selected using the orthogonal forward regression supervised feature selection method. Cognitive load was best predicted using a combination of prefrontal theta power, broad spatial range gamma power, fronto-central beta power, and fronto-central alpha power. Generalization performance estimates for pairwise classification of task difficulty using these features reached 75% for 5 s epochs, and 85% for 30 s epochs.

show abstract

EEG classification of driver mental states by deep learning

Cited by 169 publications

References 50 publications

Classification of Drowsiness Levels Based on a Deep Spatio-Temporal Convolutional Bidirectional LSTM Network Using Electroencephalography Signals

Classification of Drowsiness Levels Based on a Deep Spatio-Temporal Convolutional Bidirectional LSTM Network Using Electroencephalography Signals

Assessment of visual fatigue under LED tunable white light with different blue components

A cognitive brain–computer interface monitoring sustained attentional variations during a continuous task

Contact Info

Product

Resources

About