# Capturing individual variation in children’s electroencephalograms during nREM sleep

## TL;DR

The study explores how children's brain activity during sleep can be used to create stable individual brain fingerprints that improve with age.

## Contribution

The study introduces a probabilistic modeling approach using Bayesian reduced-rank regression to extract stable individual fingerprints from pediatric sleep EEG data.

## Key findings

- The BRRR model successfully separated subjects and generalized across sleep stages.
- Fingerprint stability increased with the age of the subjects.
- BRRR outperformed correlation-based methods in fingerprinting across sleep stages.

## Abstract

Human brain dynamics are highly unique between individuals: functional neuroimaging studies have recently described functional features that can be used as neural fingerprints. However, the stability of these fingerprints is affected by aging and disease. As such, the stability of brain fingerprints may be a useful metric when studying normal and pathological neurodevelopment. Before examining clinically relevant deviations, the individual stability and variation of neuroimaging features across brain maturation in normally developing children need to be addressed with real clinical data. Here we applied Bayesian reduced-rank regression (BRRR) to extract low-dimensional representations of electroencephalography (EEG) power spectra measured during different non-REM sleep stages (N1 and N2) from 782 normally developing children aged between 6 weeks to 19 years. The representations learned within specific sleep stages successfully separated between subjects and generalized across sleep stages. Fingerprint stability increased with the age of the subjects. Compared to correlation-based fingerprinting methods, the BRRR model performed better, especially in fingerprinting across sleep stages, highlighting the usefulness of dimensionality reduction when the noise and signal of interest are correlated. While further studies are needed to address the possible non-linear maturation effects over developmental periods, our results demonstrate the existence of stable within-session neurofunctional fingerprints in pediatric populations.

Leaning to the intuition of the uniqueness of individual brain dynamics, cortical fingerprinting based on functional brain-imaging features has gained momentum. Most fingerprinting studies have been performed on adults and not on clinical data sets. Here, we use probabilistic modeling approach on a large pediatric sleep-EEG data set to find low-dimensional individual fingerprints that generalize across sleep stages. We demonstrate that the assumption of non-independent noise is suitable for multi-channel EEG data, providing a tool for fingerprint analysis in challenging clinical contexts.