Discovery of Dynamic Functional MRI Markers of Sleep Onset Period

Understanding the Sleep Onset Period (SOP) remains challenging despite its critical role in sleep architecture. Recent research suggests that resting-state functional MRI (rs-fMRI) and its Blood-Oxygen-Level-Dependent (BOLD) signal may offer new insights. This study investigates brain activity during SOP using BOLD oscillation and focuses on detecting drowsiness through an innovative scoring system. Forty-one healthy young volunteers underwent a 45-minute rs-fMRI scan. Drowsiness was assessed by analyzing standard MRI surveillance video, specifically measuring ocular activity. This method, originally designed for road accident prevention, is applied here for the first time in a neuroimaging study. We observed an increase in low-frequency BOLD oscillations across different drowsiness states (awake, likely drowsy, drowsy, sleep), peaking at 0.05 Hz. The Power Spectral Density of BOLD oscillation at 0.05 Hz (PSD _0.05 ) increases from the earliest sign of drowsiness and progresses until sleep. This was evident in up to 58 regions, particularly within primary sensory cortices and visual and somatomotor networks. The findings suggest that PSD _0.05 in sensory cortices and networks may be a valuable marker for improving SOP delineation. Additionally, eye closure dynamics provide an objective measure of drowsiness during rs-fMRI scans, which is crucial to addressing drowsiness confounds in resting-state data.