Inter-subject transfer learning for EEG-based mental fatigue recognition

Liu, Yisi; Lan, Zirui; Cui, Jian; Sourina, Olga; Müller‐Wittig, Wolfgang

doi:10.1016/j.aei.2020.101157

Cited by 67 publications

(50 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In a sustained-attention task, reaction-times to events are directly related with drowsiness [49]. This is also true for the driving context [36], [37], [50]. The authors of the dataset collected data from 27 participants, aged from 22 to 28 years, who enrolled in 90-minute sustained-attention driving sessions at different times on the same or different days.…”

Section: Methodsmentioning

confidence: 99%

“…Other works perform cross-session instead of cross-subject [36], but this is only useful for the same subject in future trials. To the best of our knowledge, it could only be found 4 preliminary works which performed cross-subject validation, considering the driving context [37], [38], [34], [35]. In [37], the authors perform domain adaptation, a branch of transfer learning, to adapt the data distributions of source and target so that the classification could be more efficient in a crosssubject scenario.…”

Section: Related Workmentioning

confidence: 99%

“…To the best of our knowledge, it could only be found 4 preliminary works which performed cross-subject validation, considering the driving context [37], [38], [34], [35]. In [37], the authors perform domain adaptation, a branch of transfer learning, to adapt the data distributions of source and target so that the classification could be more efficient in a crosssubject scenario. In [38], EEG features, statistics, higher order crossing, fractal dimension, hjorth, signal energy, and spectral power were extracted and combined with several classifiers, such as logistic regression, linear discriminant analysis, 1nearest neighbor, linear SVM, and naïve Bayes.…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

Cross-Subject Zero Calibration Driver’s Drowsiness Detection: Exploring Spatiotemporal Image Encoding of EEG Signals for Convolutional Neural Network Classification

Paulo

Pires

Nunes

2021

IEEE Trans. Neural Syst. Rehabil. Eng.

View full text Add to dashboard Cite

This paper explores two methodologies for drowsiness detection using EEG signals in a sustained-attention driving task considering pre-event time windows, and focusing on crosssubject zero calibration. Driving accidents are a major cause of injuries and deaths on the road. A considerable portion of those are due to fatigue and drowsiness. Advanced driver assistance systems that could detect mental states which are associated with hazardous situations, such as drowsiness, are of critical importance. EEG signals are used widely for brain-computer interfaces, as well as mental state recognition. However, these systems are still difficult to design due to very low signal-tonoise ratios and cross-subject disparities, requiring individual calibration cycles. To tackle this research domain, here, we explore drowsiness detection based on EEG signals' spatiotemporal image encoding representations in the form of either recurrence plots or gramian angular fields for deep convolutional neural network (CNN) classification. Results comparing both techniques using a public dataset of 27 subjects show a superior balanced accuracy of up to 75.87% for leave-one-out cross-validation, using both techniques, against works in the literature, demonstrating the possibility to pursue cross-subject zero calibration design.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Cross-Subject Zero Calibration Driver’s Drowsiness Detection: Exploring Spatiotemporal Image Encoding of EEG Signals for Convolutional Neural Network Classification

Paulo

Pires

Nunes

2021

IEEE Trans. Neural Syst. Rehabil. Eng.

View full text Add to dashboard Cite

show abstract

“…Liu et al [45] proposed a transfer learning-based cross-subject EEG fatigue recognition algorithm without correction. They also explored the influence of the number of EEG signal channels on algorithm accuracy and compared single and multi-channel situations.…”

Section: Feature-based Transfer Learning Methodsmentioning

confidence: 99%

A Review of Transfer Learning for EEG-Based Driving Fatigue Detection

Cui

Peng

Ozawa

et al. 2021

Communications in Computer and Information Science

View full text Add to dashboard Cite

Driver mental state detection has been playing an increasingly significant role in safe driving for decades. Electroencephalogram (EEG)-based detection methods have already been applied to improve detection performance. However, numerous problems still have not been addressed in practical applications. Specifically, most of the existing traditional methods require a large number of training data, caused by differences in cross-subject samples and cross-time of the same subject, resulting in enormous calculations and time consumption. To overcome the above limitations, transfer learning, which applies data or knowledge from the source domain to the target domain, has been widely adopted in EEG processing. This article reviews the current state of mainstream transfer learning methods and their application based on driver mental state detection. To the best of our knowledge, this is the first comprehensive review of transfer learning methods for driving fatigue detection.

show abstract

“…Considering this issue, most conventional EEG analysis models are developed subject-dependently, requiring calibration time and difficult to apply to real-world applications. To address the above problem, integrated models between subjects have also been proposed [2,3,4,5,6].…”

Section: Introductionmentioning

confidence: 99%

Mitigating Inter-Subject Brain Signal Variability FOR EEG-Based Driver Fatigue State Classification

Hwang

Park

Kim

et al. 2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

With great research advances on Brain-Computer-Interface (BCI) systems, Electroencephalography (EEG) based driver fatigue state classification models have shown its effectiveness. However, EEG signals contain large differences between individuals, making it hard to build a unified model among individuals. In this paper, we propose a subjectindependent EEG-based driver fatigue state (i.e., awake, tired, and drowsy) classification model that mitigates a performance gap between subjects. To this end, we exploit an adversarial training strategy to make our classification model misclassify the subject labels. Besides, we propose an Intersubject Feature Distance Minimization (IFDM) method that minimizes the Wasserstein distance between two different subject groups of the same class to reduce the individual performance discrepancy. Our method is also designed to enable training even if the subject labels are not sufficiently included in the EEG dataset. To demonstrate the ability of the proposed method, we conduct a drowsiness classification task on a publicly available SEED-VIG dataset. The experimental results show our model achieves the highest accuracy and the lowest individual performance variability.

show abstract

Inter-subject transfer learning for EEG-based mental fatigue recognition

Cited by 67 publications

References 38 publications

Cross-Subject Zero Calibration Driver’s Drowsiness Detection: Exploring Spatiotemporal Image Encoding of EEG Signals for Convolutional Neural Network Classification

Cross-Subject Zero Calibration Driver’s Drowsiness Detection: Exploring Spatiotemporal Image Encoding of EEG Signals for Convolutional Neural Network Classification

A Review of Transfer Learning for EEG-Based Driving Fatigue Detection

Mitigating Inter-Subject Brain Signal Variability FOR EEG-Based Driver Fatigue State Classification

Contact Info

Product

Resources

About