Speeded near infrared spectroscopy (NIRS) response detection

Area of Science:

• Neuroscience
• Biomedical Engineering
• Signal Processing

Background:

• Near Infrared Spectroscopy (NIRS) measures hemodynamic responses, which are delayed relative to neural activity.
• This temporal delay limits the real-time application of NIRS-based brain-computer interfaces (BCIs) and neurofeedback systems.

Purpose of the Study:

• To investigate methods for reducing the latency in decoding behavioral states from NIRS signals.
• To enhance the practical usability of NIRS for real-time neurotechnology.

Main Methods:

• Employed a multivariate pattern classification technique, specifically a linear support vector machine (SVM).
• Evaluated various feature spaces, including signal history, gradient, hemoglobin oxygenation levels, and spatial patterns.
• Systematically decoded behavioral states from NIRS data.

Main Results:

• Achieved a 50% reduction in the latency to decode behavioral state changes, decreasing it from 4.8 seconds to 2.4 seconds.
• Demonstrated that specific feature spaces significantly impact decoding accuracy and speed.

Conclusions:

• The proposed method significantly reduces the temporal delay in NIRS-based BCI systems.
• This advancement enhances the feasibility of using NIRS for real-time applications, improving BCIs and neurofeedback systems.