A noninvasive brain computer interface using visually-induced near-infrared spectroscopy responses
Cheng-Hsuan Chen1, Ming-Shan Ho2, Kuo-Kai Shyu2
1Department of Electrical Engineering, National Central University, Taoyuan 32001, Taiwan; Department of Electrical Engineering, Fu Jen Catholic University, New Taipei City 24205, Taiwan.
Neuroscience Letters
|August 5, 2014
Summary
This study demonstrates a brain-computer interface (BCI) using near-infrared spectroscopy (NIRS). The system achieves over 90% accuracy by detecting brain responses to visual stimuli, enabling new BCI applications.
Area of Science:
- Neuroscience
- Biomedical Engineering
- Human-Computer Interaction
Background:
- Brain-computer interfaces (BCIs) offer novel interaction methods.
- Near-infrared spectroscopy (NIRS) measures hemodynamic responses in the brain.
- Developing non-invasive and accurate BCI systems is crucial.
Purpose of the Study:
- To design and evaluate a BCI system using visually-induced near-infrared spectroscopy (NIRS) responses.
- To investigate the feasibility of distinguishing between attended and unattended visual stimuli using NIRS.
- To achieve high accuracy in a BCI paradigm based on visual evoked potentials.
Main Methods:
- Utilized visually-induced NIRS responses as the basis for a BCI system.
- Employed four distinct flickering checkerboard visual stimuli on an LCD screen.
- Implemented a random resting segment duration (15-20s) to ensure independence between flickering sequences.
- Recruited six subjects who sequentially gazed at the visual stimuli in a random order.
- Applied a simple averaging process to discern NIRS responses time-locked to visual stimulus onsets.
Main Results:
- NIRS responses were successfully time-locked to the visual stimuli onsets.
- The system demonstrated the ability to discern NIRS responses from gazed versus non-gazed targets.
- Achieved classification accuracies exceeding 90% for all six subjects.
- Accuracy improved with more averaged epochs, reaching high levels after 10 or more epochs.
Conclusions:
- Visually-induced NIRS responses can be effectively utilized to develop a high-accuracy BCI system.
- The proposed method offers a promising non-invasive approach for brain-computer interaction.
- This BCI system has potential applications in assistive technologies and neurofeedback.
