Abstract:Mixed Reality (MR) technology possesses profound and extensive potential across a multitude of domains, including, but not limited to industry, healthcare, and education. However, prolonged use of MR devices to watch stereoscopic content may lead to visual fatigue. Since visual fatigue involves multiple brain regions, our study aims to explore the topological characteristics of brain networks derived from electroencephalogram (EEG) data. Because the Phase-Locked Value (PLV) is capable of effectively measuring the phase synchronization relationship between brain regions, it was calculated between all pairs of channels in both comfort and fatigue states. Subsequently, a sparse brain network was constructed based on PLV by applying an appropriate threshold. The node properties (betweenness centrality, clustering coefficient, node efficiency) and edge properties (characteristic path length) were calculated based on the corresponding brain network within specific frequency bands for both comfort and fatigue states. In analyzing the PLV of brain connectivity in comfort and fatigue states, a notable enhancement in brain connectivity is observed within the alpha, theta, and delta frequency bands during fatigue status. By analyzing the node and edge properties of brain networks, it is evident that the mean values of these properties in the fatigue state were higher than those in the comfort state. By analyzing the node and edge properties at a local level, the average difference in betweenness centrality, clustering coefficients, and nodal efficiency across the three EEG frequency bands was computed to find significant brain regions. The main findings are as follows: Betweenness centrality primarily differs in frontal and parietal regions, with minor involvement in temporal and central regions. The clustering Coefficient mainly varies in the frontal region, with slight differences being seen in the temporal and occipital regions. Nodal efficiency primarily varies in the frontal, temporal, and central regions, with minor differences being seen in the parietal and occipital regions. Edge property analysis indicates that there is a higher occurrence of long-distance connections among brain regions during the fatigue state, which reflects a loss of synaptic transmission efficiency on a global level. Our study plays a crucial role in understanding the neural mechanisms underlying visual fatigue, potentially providing insights that could be applied to high-demand cognitive fields where prolonged use of MR devices leads to visual fatigue.

EEG-Based 3D Visual Fatigue Evaluation Using CNN

Mental fatigue recognition study based on 1D convolutional neural network and short-term ECG signals

EEG driving fatigue detection based on log-Mel spectrogram and convolutional recurrent neural networks

Investigating the neural activity of various 3D visual fatigue degrees using depth‐related visual evoked potentials

EEG spatial inter-channel connectivity analysis: A GCN-based dual stream approach to distinguish mental fatigue status

Linking Attention-Based Multiscale CNN With Dynamical GCN for Driving Fatigue Detection

Attention-Guided Multi-scale Convolutional Neural Network for Driving Fatigue Detection

Multilevel Feature Fusion With 3D Convolutional Neural Network for EEG-Based Workload Estimation

Two-dimensional phase lag index image representation of electroencephalography for automated recognition of driver fatigue using convolutional neural network

Remote Photoplethysmography and Motion Tracking Convolutional Neural Network with Bidirectional Long Short-Term Memory: Non-Invasive Fatigue Detection Method Based on Multi-Modal Fusion

Machine vision detection to daily facial fatigue with a nonlocal 3D attention network

Learning Spatial–Spectral–Temporal EEG Features With Recurrent 3D Convolutional Neural Networks for Cross-Task Mental Workload Assessment

3D Visual Discomfort Assessment With a Weakly Supervised Graph Convolution Neural Network Based on Inaccurately Labeled EEG

A novel deep-learning model based on τ-shaped convolutional network (τNet) with long short-term memory (LSTM) for physiological fatigue detection from EEG and EOG signals

A Multi-Model Analysis for Driving Fatigue Detection using EEG Signals

Fatigue Characterization of EEG Brain Networks Under Mixed Reality Stereo Vision

Exploring Quantitative Assessment of Cybersickness in Virtual Reality Using EEG Signals and a CNN-LSTM Network

Objective evaluation of fatigue by EEG spectral analysis in steady-state visual evoked potential-based brain-computer interfaces

Fatigue driving detection method based on Time-Space-Frequency features of multimodal signals

Exploration of Eye Fatigue Detection Features and Algorithm Based on Eye-Tracking Signal

Identification of mental fatigue in language comprehension tasks based on EEG and deep learning