Research for estimating direction of saccadic eye movements by single trial processing

Area of Science:

• Neuroscience
• Biomedical Engineering
• Signal Processing

Background:

• Electroencephalogram (EEG) signals related to eye movements (saccades) are crucial for developing brain-computer interfaces (BCIs).
• Traditional ensemble averaging methods are unsuitable for real-time BCI applications processing raw EEG data.
• Developing novel BCI systems requires efficient real-time processing of saccade-related EEG signals.

Purpose of the Study:

• To develop a novel brain-computer interface (BCI) system utilizing electroencephalogram (EEG) signals generated during eye movements (saccades).
• To process raw EEG data in real-time using a non-conventional method, Fast ICA with Reference signal (FICAR).
• To compare saccade-related EEG signals and Independent Components (ICs) derived via FICAR in visually and auditorily guided saccade tasks.

Main Methods:

• Conducted saccade-related EEG experiments involving visually and auditorily guided saccade tasks.
• Processed recorded EEG data using Fast ICA with Reference signal (FICAR).
• Compared saccade-related EEG signals and FICAR-derived ICs based on latency and peak amplitude timing.

Main Results:

• Peak amplitude time for saccade-related ICs processed by FICAR occurred earlier than for raw EEG signals.
• This earlier peak time presents a significant advantage for developing real-time BCIs.
• Signal-to-noise (S/N) ratio improvement using FICAR was not superior to ensemble averaging.
• FICAR was subsequently used to estimate saccade direction from raw EEG signals.

Conclusions:

• The FICAR method offers a promising approach for real-time BCI development due to the earlier detection of saccade-related neural activity.
• Further research is needed to improve the S/N ratio when using FICAR for EEG signal processing.
• The ability to estimate saccade direction from raw EEG signals using FICAR opens new avenues for BCI applications.