Related Experiment Video
Updated: May 15, 2025
06:37
Electroencephalography Network Indices as Biomarkers of Upper Limb Impairment in Chronic Stroke
Published on: July 14, 2023
789
EEG-based multi-band functional connectivity using corrected amplitude envelope correlation for identifying
Jichi Chen1, Yujie Wang1, Yuguo Cui1
1School of Mechanical Engineering, Shenyang University of Technology, Shenyang, Liaoning, China.
Computer Methods in Biomechanics and Biomedical Engineering
|April 10, 2025
Summary
This study introduces a new method using Electroencephalography (EEG) and functional connectivity to detect unfavorable driving states (UDS). The approach significantly improves accuracy in recognizing UDS, enhancing driver safety.
Area of Science:
- Neuroscience
- Transportation Safety
- Biomedical Engineering
Background:
- Unfavorable driving states (UDS) pose significant risks, necessitating accurate detection methods.
- Existing functional connectivity approaches for UDS recognition face challenges due to spurious synchronization.
- Electroencephalography (EEG) signals offer a potential avenue for monitoring driver states.
Purpose of the Study:
- To develop a novel functional connectivity matrix construction approach combined with an ensemble algorithm for identifying drivers' UDS.
- To overcome the limitations of spurious synchronization in traditional functional connectivity methods.
- To enhance the accuracy and reliability of EEG-based UDS detection.
Main Methods:
- EEG data collected from a simulated driving experiment.
- Functional connectivity matrix construction using amplitude envelope correlation with leakage correction (AEC-c) across multiple frequency bands.
- Ensemble algorithm employing random subspace for k-nearest neighbors (KNN) classification.
Main Results:
- Regional AEC-c values were significantly lower for UDS compared to non-unfavorable driving states (NUDS) across beta, gamma, and all frequency bands.
- The proposed AEC-c-based functional connectivity analysis combined with random subspace KNN achieved a highest accuracy (ACC) of 96.88%.
- Performance was validated using confusion matrix, ACC, sensitivity, specificity, precision, and ROC curves with 5-fold cross-validation.
Conclusions:
- The novel AEC-c-based functional connectivity framework effectively identifies drivers' UDS from EEG signals.
- This method offers a robust solution to spurious synchronization issues in functional connectivity analysis.
- The findings contribute to improved man-machine systems and enhanced road safety through reliable EEG-based driver monitoring.
