Effectiveness Evaluation of Real-Time Scalp Signal Separating Algorithm on Near-Infrared Spectroscopy Neurofeedback

Area of Science:

• Neuroscience
• Biomedical Engineering
• Signal Processing

Background:

• Near-infrared spectroscopy (NIRS) measures brain activity via hemoglobin concentration changes.
• Superficial hemodynamic signals from the scalp can contaminate NIRS data, impacting neurofeedback and brain-computer interface (BCI) applications.
• Real-time processing is crucial for removing these interfering scalp signals.

Purpose of the Study:

• To evaluate the effectiveness of the real-time scalp signal separating (RT-SSS) algorithm in improving NIRS-based neurofeedback.
• To compare neurofeedback performance using raw NIRS signals versus NIRS signals processed by the RT-SSS algorithm.

Main Methods:

• Two neurofeedback experiments were conducted using NIRS.
• Experiment 1 utilized raw NIRS signals for feedback.
• Experiment 2 used deep NIRS signals extracted by the RT-SSS algorithm for feedback.
• Participants controlled a feedback signal to follow a target track; accuracy was measured.

Main Results:

• Neurofeedback performance was assessed using accuracy scores comparing controlled and target tracks.
• The experiment employing the RT-SSS algorithm (Experiment 2) demonstrated superior accuracy compared to using raw NIRS signals (Experiment 1).

Conclusions:

• The RT-SSS algorithm effectively separates scalp-blood signals from NIRS data.
• Applying the RT-SSS algorithm significantly benefits neurofeedback applications by enhancing signal quality and user performance.