DTU BCI speller: an SSVEP-based spelling system with dictionary support

Area of Science:

• Neuroscience
• Biomedical Engineering
• Human-Computer Interaction

Background:

• Brain-computer interfaces (BCIs) offer alternative communication methods for individuals with severe motor impairments.
• Steady-state visual evoked potential (SSVEP) based BCIs are known for their high information transfer rates.
• Existing SSVEP BCIs often require complex setups and extensive training, limiting user-friendliness.

Purpose of the Study:

• To introduce and evaluate the DTU BCI Speller, a novel SSVEP-based BCI system.
• To assess the system's simplicity, user-friendliness, and performance in terms of information transfer rate (ITR) and characters per minute (CPM).
• To investigate the impact of dictionary support on sentence writing efficiency.

Main Methods:

• Development of the DTU BCI Speller utilizing a single electrode for SSVEP signal acquisition.
• Stimulus presentation on a liquid crystal display (LCD).
• Inclusion of dictionary support to aid in sentence construction.
• Performance evaluation with nine healthy subjects writing full sentences.

Main Results:

• The DTU BCI Speller achieved an average ITR of 21.94 ± 15.63 bits/min.
• Average characters per minute (CPM) was 4.90 ± 3.84, with a maximum of 8.74 CPM.
• Subjects reported high user-friendliness, and dictionary support significantly reduced sentence writing time for proficient users.

Conclusions:

• The DTU BCI Speller demonstrates significant potential as a user-friendly and efficient brain-computer interface for text communication.
• The system's simplicity (single electrode) and SSVEP-based approach contribute to its accessibility.
• Dictionary support enhances usability and speed, particularly for users with higher classification accuracy.