Driver drowsiness estimation using EEG signals with a dynamical encoder\u2013decoder modeling framework

Sadegh Arefnezhad,Matthias Frühwirth,Ioana Victoria Koglbauer,Ali Yousefi,Anja Ischebeck,Maximilian Moser,James Hamet,Arno Eichberger

doi:10.1038/s41598-022-05810-x

Abstract

Drowsiness is a leading cause of accidents on the road as it negatively affects the driver’s ability to safely operate a vehicle. Neural activity recorded by EEG electrodes is a widely used physiological correlate of driver drowsiness. This paper presents a novel dynamical modeling solution to estimate the instantaneous level of the driver drowsiness using EEG signals, where the PERcentage of eyelid CLOSure (PERCLOS) is employed as the ground truth of driver drowsiness. Applying our proposed modeling framework, we find neural features present in EEG data that encode PERCLOS. In the decoding phase, we use a Bayesian filtering solution to estimate the PERCLOS level over time. A data set that comprises 18 driving tests, conducted by 13 drivers, has been used to investigate the performance of the proposed framework. The modeling performance in estimation of PERCLOS provides robust and repeatable results in tests with manual and automated driving modes by an average RMSE of 0.117 (at a PERCLOS range of 0 to 1) and average High Probability Density percentage of 62.5%. We further hypothesized that there are biomarkers that encode the PERCLOS across different driving tests and participants. Using this solution, we identified possible biomarkers such as Theta and Delta powers. Results show that about 73% and 66% of the Theta and Delta powers which are selected as biomarkers are increasing as PERCLOS grows during the driving test. We argue that the proposed method is a robust and reliable solution to estimate drowsiness in real-time which opens the door in utilizing EEG-based measures in driver drowsiness detection systems.

Highlights

Introduction to StatisticalLearning, vol 103 of Springer Texts in Statistics, 59–126, https://doi.org/10.1007/978-1-4614-7138-7_3 (Springer 2013). 56
We searched across all EEG features to identify strong correlations to PERcentage of eyelid CLOSure (PERCLOS) based on their slope
This means that regardless of the user, these features are significantly realted to the PERCLOS values recorded during the driving test

Summary

Introduction

Introduction to StatisticalLearning, vol 103 of Springer Texts in Statistics, 59–126, https://doi.org/10.1007/978-1-4614-7138-7_3 (Springer 2013). 56. Vol 103 of Springer Texts in Statistics, 59–126, https://doi.org/10.1007/978-1-4614-7138-7_3 (Springer 2013). A. Feature extraction of epilepsy eeg using discrete wavelet transform. In 2016 12th International Computer Engineering Conference (ICENCO), 190–195, https://doi.org/10.1109/ICENCO.2016.78564 67 (IEEE, 12/28/2016 - 12/29/2016). Single-channel-based automatic drowsiness detection architecture with a reduced number of EEG features. S. Classification of EEG signals based on pattern recognition approach. S. et al Support vector machine based detection of drowsiness using minimum EEG features. In 2013 International Conference on Social Computing, 827–835, https://doi.org/10.1109/SocialCom.2013.124 (2013). Driver cognitive workload estimation: A data-driven perspective. Driver workload estimation using a novel hybrid method of error reduction ratio causality and support vector machine.

Methods

Results

Discussion

Conclusion