Relationship between amplitude of resting-state fNIRS global signal and EEG vigilance measures.

Most of the prior studies of functional connectivity in both healthy and diseased brain utilized resting-state functional magnetic resonance imaging (fMRI) as a measure to represent the temporal synchrony in blood oxygenation level dependent (BOLD) signals across brain regions. To eliminate the impact of widely distributed global signal component across the brain, many studies have adopted global signal regression (GSR) as a pre-processing approach to regress the global signal component out of BOLD signals followed by computing hemodynamic connectivity. However, the procedure of global signal regression has been debated as physiologically relevant component may be present in global signal. In this study, we aimed to address the controversy of global signal using functional non-invasive neuroimaging technology, i.e. functional near-infrared spectroscopy (fNIRS), which measures hemodynamic signals by probing local changes in oxygen consumption, a common imaging contrast measured by BOLD fMRI. In the current study, we acquired simultaneous EEG and fNIRS signals, both in high-density configuration and whole-brain coverage, in healthy individuals at eyes-open and eyes-closed resting state and at three different body positions. We explored the underlying relationship between fNIRS global signal and EEG vigilance, and have identified negative correlation between fNIRS global signal and EEG vigilance across the physiological variations of measurements.

EEG fluctuations of wake and sleep in mild cognitive impairment.

Amnestic Mild Cognitive Impairment (aMCI), a condition in which the memory functions of cognition are significantly impaired, is an established risk factor for Alzheimer's disease. Electroencephalography (EEG) is a tool capable of measuring the dynamics of the brain's neural networks, and is thus an important means in analysis and understanding of aMCI. In this proof-of-concept study, we compared the brain activation patterns of ten aMCI subjects with those of four healthy subjects during sleep by employing a 64-channel EEG data collection system. The power spectrum was analyzed to identify sleep stages, while spectral topography and source imaging techniques were employed to study the fluctuating patterns of the brain. Results of this study show an increase in activation power across all sleep stages in the delta and theta frequency bands alongside a decrease in alpha band activity for aMCI subjects. Source imaging analysis of the resting EEG identified default mode network, which becomes decoupled as sleep stages deepen. In the proof-of-concept study, our exploratory analysis demonstrated the feasibility of imaging dynamic network organization using EEG in aMCI.