Evoked Hemodynamic Response Estimation to Auditory Stimulus Using Recursive Least Squares Adaptive Filtering with...

Area of Science:

• Neuroscience
• Biomedical Engineering
• Signal Processing

Background:

• Functional near-infrared spectroscopy (fNIRS) performance is often compromised by physiological interference.
• Sources of interference include cardiac pulse, breathing, and low-frequency oscillations.
• Previous work proposed recursive least squares (RLS) adaptive filtering for interference reduction using simulations.

Purpose of the Study:

• To evaluate the effectiveness of RLS adaptive filtering on human data for detecting evoked hemodynamic responses.
• To assess the suitability of multidistance fNIRS measurements combined with RLS filtering for brain functional activation studies.
• To validate RLS adaptive filtering as a practical tool for real-time brain activity detection.

Main Methods:

• Utilized a multidistance probe with continuous wave fNIRS for data acquisition.
• Applied recursive least squares (RLS) adaptive filtering to suppress physiological interference.
• Analyzed fNIRS data to detect hemodynamic responses to auditory stimuli.

Main Results:

• RLS adaptive filtering demonstrated effectiveness in suppressing global physiological interference in human fNIRS data.
• The study successfully detected evoked hemodynamic responses to auditory stimuli.
• Multidistance fNIRS measurements combined with RLS filtering proved practical for brain activity detection.

Conclusions:

• RLS adaptive filtering is a validated and effective method for reducing physiological noise in fNIRS.
• Multidistance fNIRS measurements coupled with RLS filtering offer a practical approach for real-time brain activity monitoring.
• This approach enhances the reliability of fNIRS for studying brain functional activation.