A Modular Framework for EEG Web Based Binary Brain Computer Interfaces to Recover Communication Abilities in Impaired...

Area of Science:

• Neuroscience
• Computer Science
• Rehabilitation Engineering

Background:

• Brain Computer Interfaces (BCIs) offer a vital communication channel for individuals with severe motor impairments.
• Electroencephalography (EEG) is a practical neuroimaging technique for implementing BCIs.
• Existing BCI systems can be slow and lack personalization, hindering user engagement and efficiency.

Purpose of the Study:

• To develop a modular framework for binary BCIs to facilitate communication and reduce text writing time.
• To incorporate a motivational tool to enhance EEG signal quality.
• To integrate a predictive engine for faster text input based on language statistics.

Main Methods:

• A modular framework was designed encompassing signal acquisition, analysis, classification, communication, visualization, and a predictive engine.
• The framework supports a graphic interface for symbol selection in a table.
• A predictive module utilizes letter and word frequencies for enhanced text generation.

Main Results:

• The modular framework allows for easy customization of the graphic interface for individual BCI users.
• The system demonstrated potential in reducing text writing time through its predictive capabilities.
• Experimental results on healthy subjects provide a foundation for future development and application in clinical settings.

Conclusions:

• The proposed modular BCI framework is adaptable and can be integrated with various classification strategies and communication paradigms.
• The system's design, including motivational and predictive elements, aims to improve BCI performance and user experience.
• This framework serves as a valuable starting point for developing advanced BCIs for severely disabled individuals.