Improving EEG-Based Motor Imagery Classification for Real-Time Applications Using the QSA Method

Area of Science:

• Neuroscience
• Signal Processing
• Biomedical Engineering

Background:

• Electroencephalography (EEG) is crucial for understanding brain activity.
• Real-time analysis of EEG signals, especially for motor imagery (IM), presents significant challenges.
• Existing quaternion-based signal analysis (QSA) methods require improvement for efficiency and accuracy.

Purpose of the Study:

• To introduce an improved quaternion-based signal analysis (iQSA) methodology.
• To enhance the extraction of EEG signal features for real-time applications.
• To improve classification accuracy and efficiency in motor imagery tasks.

Main Methods:

• Developed the improved quaternion-based signal analysis (iQSA) technique.
• Extracted EEG signal features including average, variance, homogeneity, and contrast.
• Implemented boosting-technique-based decision trees for classification.
• Investigated the impact of variable sampling periods (0.5s to 3s) on classification effectiveness.

Main Results:

• iQSA demonstrated significantly improved classification accuracy compared to the original QSA technique.
• Achieved an 82.30% accuracy rate with 0.5s samples and 73.16% with 3s samples.
• The original QSA technique yielded accuracy rates between 33.31% and 41.07%.

Conclusions:

• The iQSA technique offers superior performance for real-time EEG analysis.
• iQSA is more efficient, requiring fewer samples for accurate signal classification.
• The findings support the suitability of iQSA for developing advanced real-time applications in motor imagery.