UC Irvine

High-frequency oscillations (HFOs) are short bursts of power at frequencies > 80 Hz, and they are thought to be significant markers of both cognition and disease in humans and animals. Physiological HFOs have been associated with basic vision and motor, and memory consolidation processes in humans. There is also a significant interest in high-frequency oscillations as a biomarker for epileptogenicity. Pathological HFOs have been associated with many different types of epilepsy and seen in patients of all ages. While studies of epilepsy are mainly associated with pathological HFOs, these studies can also benefit from studying physiological HFOs.

The initial studies of HFOs were conducted using intracranial electroencephalogram (EEG) recordings; however, there is recent evidence that they can also be detected in scalp EEG. While several studies have successfully used scalp EEG to study high-frequency oscillations, they have primarily focused on epilepsy and pathological HFOs. In contrast,spontaneously occurring physiological HFOs in healthy human subjects using scalp EEG have received little attention thus far. Therefore, the goal of our study is to measure physiological ripples in healthy infants using scalp EEG and obtain robust estimates of their spatiotemporal characteristics.

Here, we report the detection of spontaneously occurring physiological ripples in the long-term scalp EEG of healthy infant subjects. Events were automatically detected in all sixteen subjects and confirmed via visual validation. In total, 11,771 visually validated HFOs were analyzed. We characterized their duration, peak frequency, root-mean-square (RMS) amplitude, spatial distribution, global HFO rate, and variability of global HFO rate across sleep stages and over time. We found that HFO rate was highest in frontal and temporal channels, and it was highest in the lightest stage of non-REM sleep (N1) across all subjects. Based on 10-minute segments of EEG, the measurements of rate varied over time, with the highest variance in stage N1. We found no relationship between subject age and global HFO rate.

This work represents the most comprehensive analysis of scalp physiological ripples thus far, drawing from almost 180 hours of non-REM sleep EEG. The results contribute to our understanding of the visibility and characteristics of physiological ripples on the scalp and their relationship to the stages of sleep, as well as providing a valuable baseline for studies of pathological ripples associated with epilepsy.