Efficient Generalized Electroencephalography-Based Drowsiness Detection Approach with Minimal Electrodes

Abstract

Drowsiness is a main factor for various costly defects, even fatal accidents in areas such as construction, transportation, industry and medicine, due to the lack of monitoring vigilance in the mentioned areas. The implementation of a drowsiness detection system can greatly help to reduce the defects and accident rates by alerting individuals when they enter a drowsy state. This research proposes an electroencephalography (EEG)-based approach for detecting drowsiness. EEG signals are passed through a preprocessing chain composed of artifact removal and segmentation to ensure accurate detection followed by different feature extraction methods to extract the different features related to drowsiness. This work explores the use of various machine learning algorithms such as Support Vector Machine (SVM), the K nearest neighbor (KNN), the Naive Bayes (NB), the Decision Tree (DT), and the Multilayer Perceptron (MLP) to analyze EEG signals sourced from the DROZY database, carefully labeled into two distinct states of alertness (awake and drowsy). Segmentation into 10 s intervals ensures precise detection, while a relevant feature selection layer enhances accuracy and generalizability. The proposed approach achieves high accuracy rates of 99.84% and 96.4% for intra (subject by subject) and inter (cross-subject) modes, respectively. SVM emerges as the most effective model for drowsiness detection in the intra mode, while MLP demonstrates superior accuracy in the inter mode. This research offers a promising avenue for implementing proactive drowsiness detection systems to enhance occupational safety across various industries.

Keywords: EEG signals; drowsiness detection; feature selection; machine learning.

Figure 1

Drowsiness detection with EEG signals general processing chain.

Figure 2

DROZY EEG signals [52]: (a) location of EEG electrodes according to international system 10–20 (Fz, Cz, C3, C4, Pz); (b) EEG raw.

Figure 3

DROZY EEG signals for alert subject.

Figure 4

DROZY EEG signals for drowsy subject.

Figure 5

The general scheme of the proposed method.

Figure 6

EEG signal decomposition by the wavelet function [39].

Figure 7

Principle of operation of the RFECV.

Figure 8

Data distribution for the train and test sets in the intra mode.

Figure 9

Confusion matrix. True Positive (TP): prediction of drowsiness when the actual state is drowsiness. False Positive (FP): prediction of drowsiness when the real state is alertness. True Negative (TN): prediction of alertness when the real state is alertness. False Negative (FN): prediction of alertness when the real state is drowsiness.

Figure 10

Classification accuracy with 30 and 10 s segments.

Figure 11

Different classifiers’ accuracy after RFECV.

Figure 12

The number of features selected by RFECV with SVM.

Figure 13

SVM train and test accuracy.

Figure 14

Different classifiers’ accuracy with different EEG deviations in the inter mode.

Figure 15

The number of features selected by RFECV with MLP.

Figure 16

MLP train and test accuracy.

Figure 17

The effect of the feature selection method on the accuracy of the approach.