Spatiotemporal and time-frequency analysis of functional near infrared spectroscopy brain signals using independent component analysis

Abstract

Functional near infrared spectroscopy (fNIRS) is a noninvasive method to capture brain activities according to the measurements of changes in both oxyhemoglobin and deoxyhemoglobin concentrations. However, fNIRS recordings are the hemodynamic signals that come from the latent neural sources that are spatially and temporally mixed across the brain. The purpose of this work is to extract the temporal and frequency characteristics as well as the spatial activation patterns in the brains using independent component analysis (ICA). In this study, the filtered fNIRS recordings were processed and the time-frequency and spatiotemporal domain independent components (ICs) were identified by ICA. We found that multiple task-related components can be separated by ICA in time-frequency domain, and distinct spatial patterns of brain activity can be derived from ICs that are well correlated with the specific neural events, such as finger tapping tasks.