A Brain Network Analysis Model for Motion Sickness in Electric Vehicles Based on EEG and fNIRS Signal Fusion

Area of Science:

• Neuroscience
• Biomedical Engineering
• Transportation Technology

Background:

• Motion sickness is a prevalent issue in electric vehicles, diminishing passenger experience.
• Current methods for assessing motion sickness lack objective, real-time measurement capabilities.

Purpose of the Study:

• To develop and validate a functional brain network analysis model for detecting motion sickness.
• To integrate electroencephalography (EEG) and functional near-infrared spectroscopy (fNIRS) for enhanced motion sickness detection.

Main Methods:

• Collected EEG and fNIRS data from 32 participants during real-world electric vehicle testing.
• Analyzed EEG signals across delta, theta, alpha, and beta frequency bands.
• Calculated brain oxygenation variation from fNIRS and constructed functional brain network models.
• Developed a graph convolutional network (GCN) model to integrate multimodal neuroimaging data.

Main Results:

• Significant differences in brain functional connectivity were observed between motion sickness and non-motion sickness states.
• The integrated EEG-fNIRS model demonstrated superior performance compared to single-modality approaches.
• The combined model improved the F1 score by 11.4% over EEG alone and 8.2% over fNIRS alone.

Conclusions:

• Integrating EEG and fNIRS signals with GCN offers a robust method for motion sickness detection.
• This multimodal approach significantly enhances detection accuracy over individual methods.
• The developed model shows strong potential for practical application in mitigating motion sickness in electric vehicles.