# NEURAL: quantitative features for newborn EEG using Matlab

arXiv: https://arxiv.org/abs/1704.05694 · HTML: https://mlsanity.com/paper/1704.05694/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 19 Apr 2017.

## Built on by (most cited first, 3 of 3)

- [Neonatal EEG graded for severity of background abnormalities in hypoxic-ischaemic encephalopathy](https://mlsanity.com/paper/2206.04420/) (2022)
- [Machine-learning competition to grade EEG background patterns in newborns with hypoxic-ischaemic encephalopathy](https://mlsanity.com/paper/2509.09695/) (2025)
- [Expert-level detection of low-voltage intervals in EEG as a robust biomarker of neonatal encephalopathy](https://mlsanity.com/paper/2610.12680/) (2026)

## Abstract (arXiv)

Background: For newborn infants in critical care, continuous monitoring of brain function can help identify infants at-risk of brain injury. Quantitative features allow a consistent and reproducible approach to EEG analysis, but only when all implementation aspects are clearly defined. Methods: We detail quantitative features frequently used in neonatal EEG analysis and present a Matlab software package together with exact implementation details for all features. The feature set includes stationary features that capture amplitude and frequency characteristics and features of inter-hemispheric connectivity. The software, a Neonatal Eeg featURe set in mAtLab (NEURAL), is open source and freely available. The software also includes a pre-processing stage with a basic artefact removal procedure. Conclusions: NEURAL provides a common platform for quantitative analysis of neonatal EEG. This will support reproducible research and enable comparisons across independent studies. These features present summary measures of the EEG that can also be used in automated methods to determine brain development and health of the newborn in critical care.