Neuronal Activation Detection Using Vector Phase Analysis with Dual Threshold Circles: A Functional Near-Infrared...

Area of Science:

• Neuroscience
• Biomedical Engineering
• Signal Processing

Background:

• Functional near-infrared spectroscopy (fNIRS) measures brain activity via hemoglobin changes.
• Distinguishing initial dip and hemodynamic response (HR) phases is crucial for accurate interpretation.
• Existing methods may not effectively differentiate these phases.

Purpose of the Study:

• To develop a novel vector phase diagram for differentiating the initial dip and HR phases in fNIRS signals.
• To apply this diagram to a brain-computer interface (BCI) for classifying motor tasks.
• To evaluate the performance improvement offered by the new method.

Main Methods:

• A vector phase diagram was created using oxy-hemoglobin (HbO) and deoxy-hemoglobin (HbR) changes in polar coordinates.
• Dual threshold circles were incorporated to identify the initial dip.
• The scheme was tested on a BCI classifying right-hand thumb and little finger tapping tasks using linear discriminant analysis.

Main Results:

• The vector phase diagram effectively differentiated brain activity locations for two distinct motor tasks.
• The initial dip map showed greater spatial specificity compared to the HR map.
• Classification accuracy improved significantly from 59% to 74.9% using the phase diagram with dual threshold circles.

Conclusions:

• The developed vector phase diagram is a valuable tool for analyzing fNIRS data, particularly for distinguishing hemodynamic response components.
• This method enhances the performance of brain-computer interfaces by improving the classification of motor tasks.
• The findings suggest improved spatial resolution and accuracy in brain activity localization using the initial dip feature.