Multi-source signal alignment and efficient multi-dimensional feature classification in the application of EEG-based subject-independent drowsiness detection

Shen, Mu; Zou, Bing; Li, Xinhang; Zheng, Yubo; Li, Lei; Zhang, Lin

doi:10.1016/j.bspc.2021.103023

Cited by 14 publications

(6 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Predictor importance-based feature selection methods greatly optimize the feature extraction process by selecting only the promising and important features which are best representative of the response variable. Cui et al (2021 ), Shen et al (2021 ), Cui et al (2022 ), and Kim et al (2022) used DL models for feature extraction and classification, which incur higher computational costs, but they used spectral features, which are more effective and representative of physiological brain states. On the other hand, Awais et al (2017 ), Wei et al (2018 ), and Choi et al (2019) used self-extracted spectral features and conventional ML classifiers for drowsiness detection, similar to this study.…”

Section: Discussionmentioning

confidence: 99%

“…The minimum execution time of the overall scheme also resulted from reducing the number of channels required to be processed for activity classification. Shen et al (2021 ) and Kim et al (2022) used ≥30 EEG channels from multiple cortices, while other studies mentioned in Table 3 used fewer EEG channels (1–7) from selected cortices of the brain (PFC, FC, OC, TC, and mastoids). It is to be noted that all the studies have included OC channels (O1, O2, Oz) in their work, while FC channels (F7, F8) are among the selected channels in Wei et al (2018) , Choi et al (2019) and this study.…”

Section: Discussionmentioning

confidence: 99%

“…These works are similar and comparable as they have presented the offline driver drowsiness detection scheme based on physiological data collected during the lane-keeping task in a simulated driving environment.The feature extraction process significantly increases the processing time and computational cost during real-time classification tasks when a large number of features are involved. Predictor importance-based feature selection methods greatly optimize the feature extraction process by selecting only the promising and important features which are best representative of the response variable Cui et al (2021),Shen et al (2021). Cui et al (2022), and Kim et al (…”

mentioning

confidence: 99%

See 2 more Smart Citations

Driving drowsiness detection using spectral signatures of EEG-based neurophysiology

Arif

Munawar²,

Ali³

2023

Front. Physiol.

View full text Add to dashboard Cite

Introduction: Drowsy driving is a significant factor causing dire road crashes and casualties around the world. Detecting it earlier and more effectively can significantly reduce the lethal aftereffects and increase road safety. As physiological conditions originate from the human brain, so neurophysiological signatures in drowsy and alert states may be investigated for this purpose. In this preface, A passive brain-computer interface (pBCI) scheme using multichannel electroencephalography (EEG) brain signals is developed for spatially localized and accurate detection of human drowsiness during driving tasks.Methods: This pBCI modality acquired electrophysiological patterns of 12 healthy subjects from the prefrontal (PFC), frontal (FC), and occipital cortices (OC) of the brain. Neurological states are recorded using six EEG channels spread over the right and left hemispheres in the PFC, FC, and OC of the sleep-deprived subjects during simulated driving tasks. In post-hoc analysis, spectral signatures of the δ, θ, α, and β rhythms are extracted in terms of spectral band powers and their ratios with a temporal correlation over the complete span of the experiment. Minimum redundancy maximum relevance, Chi-square, and ReliefF feature selection methods are used and aggregated with a Z-score based approach for global feature ranking. The extracted drowsiness attributes are classified using decision trees, discriminant analysis, logistic regression, naïve Bayes, support vector machines, k-nearest neighbors, and ensemble classifiers. The binary classification results are reported with confusion matrix-based performance assessment metrics.Results: In inter-classifier comparison, the optimized ensemble model achieved the best results of drowsiness classification with 85.6% accuracy and precision, 89.7% recall, 87.6% F1-score, 80% specificity, 70.3% Matthews correlation coefficient, 70.2% Cohen’s kappa score, and 91% area under the receiver operating characteristic curve with 76-ms execution time. In inter-channel comparison, the best results were obtained at the F8 electrode position in the right FC of the brain. The significance of all the results was validated with a p-value of less than 0.05 using statistical hypothesis testing methods.Conclusions: The proposed scheme has achieved better results for driving drowsiness detection with the accomplishment of multiple objectives. The predictor importance approach has reduced the feature extraction cost and computational complexity is minimized with the use of conventional machine learning classifiers resulting in low-cost hardware and software requirements. The channel selection approach has spatially localized the most promising brain region for drowsiness detection with only a single EEG channel (F8) which reduces the physical intrusiveness in normal driving operation. This pBCI scheme has a good potential for practical applications requiring earlier, more accurate, and less disruptive drowsiness detection using the spectral information of EEG biosignals.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Driving drowsiness detection using spectral signatures of EEG-based neurophysiology

Arif

Munawar²,

Ali³

2023

Front. Physiol.

View full text Add to dashboard Cite

show abstract

“…The extracted features in the HE-based method for drowsiness detection fall into three main domains: time [19][20][21], frequency [11,22], and spatial domains. Moreover, multidomain features have also been employed in HE-based methods.…”

Section: Hand-engineered (He) Feature-based Methodsmentioning

confidence: 99%

EEG_DMNet: A Deep Multi-Scale Convolutional Neural Network for Electroencephalography-Based Driver Drowsiness Detection

Obaidan,

Hussain,

AlMajed

2024

Electronics

View full text Add to dashboard Cite

Drowsy driving is one of the major causes of traffic accidents, injuries, and deaths on roads worldwide. One of the best physiological signals that are useful in detecting a driver’s drowsiness is electroencephalography (EEG), a kind of brain signal that directly measures neurophysiological activities in the brain and is widely utilized for brain–computer interfaces (BCIs). However, designing a drowsiness detection method using EEG signals is still challenging because of their non-stationary nature. Deep learning, specifically convolutional neural networks (CNNs), has recently shown promising results in driver’s drowsiness. However, state-of-the-art CNN-based methods extract features sequentially and discard multi-scale spectral-temporal features, which are important in tackling the non-stationarity of EEG signals. This paper proposes a deep multi-scale convolutional neural network (EEG_DMNet) for driver’s drowsiness detection that learns spectral-temporal features. It consists of two main modules. First, the multi-scale spectral-temporal features are extracted from EEG trials using 1D temporal convolutions. Second, the spatial feature representation module calculates spatial patterns from the extracted multi-scale features using 1D spatial convolutions. The experimental results on the public domain benchmark SEED-VIG EEG dataset showed that it learns discriminative features, resulting in an average accuracy of 97.03%, outperforming the state-of-the-art methods that used the same dataset. The findings demonstrate that the proposed method effectively and efficiently detects drivers’ drowsiness based on EEG and can be helpful for safe driving.

show abstract

“…In recent years, deep transfer learning based on domain adaptation has shown the advantage of a high recognition rate [ 29 – 31 ], particularly regarding MI-EEG based BCI rehabilitation systems [ 32 – 38 ]. Jeon et al [ 32 ] devised a multi-path network framework for MI classification, which realized domain adaptation by adapting samples of other subjects and using the gradient reversal layer to update network parameters and improve the network performance.…”

Section: Introductionmentioning

confidence: 99%

A dual alignment-based multi-source domain adaptation framework for motor imagery EEG classification

2022

Appl Intell

View full text Add to dashboard Cite

Domain adaptation, as an important branch of transfer learning, can be applied to cope with data insufficiency and high subject variabilities in motor imagery electroencephalogram (MI-EEG) based brain-computer interfaces. The existing methods generally focus on aligning data and feature distribution; however, aligning each source domain with the informative samples of the target domain and seeking the most appropriate source domains to enhance the classification effect has not been considered. In this paper, we propose a dual alignment-based multi-source domain adaptation framework, denoted DAMSDAF. Based on continuous wavelet transform, all channels of MI-EEG signals are converted respectively and the generated time-frequency spectrum images are stitched to construct multi-source domains and target domain. Then, the informative samples close to the decision boundary are found in the target domain by using entropy, and they are employed to align and reassign each source domain with normalized mutual information. Furthermore, a multi-branch deep network (MBDN) is designed, and the maximum mean discrepancy is embedded in each branch to realign the specific feature distribution. Each branch is separately trained by an aligned source domain, and all the single branch transfer accuracies are arranged in descending order and utilized for weighted prediction of MBDN. Therefore, the most suitable number of source domains with top weights can be automatically determined. Extensive experiments are conducted based on 3 public MI-EEG datasets. DAMSDAF achieves the classification accuracies of 92.56%, 69.45% and 89.57%, and the statistical analysis is performed by the kappa value and t- test. Experimental results show that DAMSDAF significantly improves the transfer effects compared to the present methods, indicating that dual alignment can sufficiently use the different weighted samples and even source domains at different levels as well as realizing optimal selection of multi-source domains.

show abstract

Multi-source signal alignment and efficient multi-dimensional feature classification in the application of EEG-based subject-independent drowsiness detection

Cited by 14 publications

References 26 publications

Driving drowsiness detection using spectral signatures of EEG-based neurophysiology

Driving drowsiness detection using spectral signatures of EEG-based neurophysiology

EEG_DMNet: A Deep Multi-Scale Convolutional Neural Network for Electroencephalography-Based Driver Drowsiness Detection

A dual alignment-based multi-source domain adaptation framework for motor imagery EEG classification

Contact Info

Product

Resources

About