Thought-Actuated Wheelchair Navigation with Communication Assistance Using Statistical Cross-Correlation-Based...

Area of Science:

• Neuroscience
• Biomedical Engineering
• Human-Computer Interaction

Background:

• Investigates a novel electroencephalogram (EEG) based approach for brain-computer interface (BCI) implementation.
• Focuses on developing a thought-controlled wheelchair navigation system with communication assistance.

Purpose of the Study:

• To design and validate a BCI system for intuitive wheelchair control and communication.
• To evaluate the efficacy of EEG signal processing techniques for feature extraction and classification.

Main Methods:

• Recorded EEG signals from 10 participants performing seven distinct mental tasks.
• Processed EEG data by noise elimination, frequency band partitioning, and cross-correlation analysis for feature extraction.
• Utilized statistical measures and an online sequential-extreme learning machine (ELM) for feature validation and classification.

Main Results:

• Identified the μ (r) feature set derived from cross-correlation signals as the most effective.
• Achieved a high recognition rate of 91.93% using the optimal feature set.
• Demonstrated the feasibility of the proposed EEG-based BCI for practical applications.

Conclusions:

• The μ (r) feature set demonstrates superior performance in recognizing mental tasks for BCI control.
• The developed system offers a promising solution for enhancing mobility and communication for individuals with disabilities.
• Cross-correlation analysis combined with ELM provides an effective method for EEG-based BCI feature extraction and classification.