Classification Approach for Attention Assessment via Singular Spectrum Analysis Based on Single-Channel...

Area of Science:

• Neuroscience
• Biomedical Engineering
• Signal Processing

Background:

• Attention assessment is vital for diagnosing Attention Deficit Hyperactivity Disorder (ADHD).
• Electroencephalograms (EEGs) offer a promising modality for objective attention evaluation.
• Existing methods may lack sufficient feature extraction and noise reduction capabilities.

Purpose of the Study:

• To develop and evaluate an advanced classification approach for attention assessment using single-channel EEGs.
• To improve the accuracy of attention assessment by incorporating advanced signal processing techniques.
• To compare the performance of the proposed method against traditional approaches.

Main Methods:

• Acquisition of single-channel electroencephalograms (EEGs) during various activities.
• Application of Fast Fourier Transform (FFT) for denoising by discarding high-frequency components.
• Utilizing Empirical Mode Decomposition (EMD) to remove signal trends and Singular Spectrum Analysis (SSA) for enhanced feature extraction.
• Classification using Random Forest, Support Vector Machine (SVM), and Back-Propagation (BP) neural networks.

Main Results:

• The proposed method, incorporating EMD and SSA, demonstrated superior classification performance.
• Attention scores were derived from classification accuracy percentages.
• Numerical simulations confirmed the enhanced classification accuracy compared to methods without EMD and SSA.

Conclusions:

• The integrated signal processing technique (FFT, EMD, SSA) significantly improves EEG-based attention assessment.
• This method offers a more accurate and robust approach for attention evaluation, potentially aiding in ADHD diagnosis.
• The findings highlight the importance of advanced signal processing in extracting meaningful features from EEG data for cognitive assessments.