EEG-based deep learning estimates preterm infants’ cerebral age: Contribution of slow-wave and cross-frequency coupling
Abstract
Reliably estimating cerebral maturation in preterm newborns is crucial for early identification of atypical neurodevelopment. We developed a deep-learning framework using a convolutional neural network (CNN) to predict postmenstrual age (PMA) from spontaneous neural activity in EEG recordings of 165 healthy preterm infants (28–38 weeks PMA). The model achieved a mean absolute error of 0.75 ± 0.60 weeks, with 73.9% of prediction errors within one week. To identify key neurophysiological drivers, we systematically manipulated signal components. Results showed that removing low-frequency activity (<4 Hz) significantly degraded performance and disrupting phase-amplitude coupling selectively impaired prediction accuracy before 34 weeks PMA, while high-frequency activity (>12 Hz) had minimal impact. Our results demonstrate that maturation signatures are primarily driven by slow-wave dynamics and their hierarchical coupling with faster oscillations, which are fundamental hallmarks of early neurodevelopment.
Related
- Cross-Frequency Coupling as a Neural Substrate for Prediction Error Evaluation: A Laminar Neural Mass Modeling Approach
- An integrated machine learning framework for cerebral aneurysm rupture status classification using clinical, morphological, and hemodynamic features
- Deep learning-based prediction of wall shear stress throughout the cardiac cycle in middle cerebral artery bifurcation aneurysms
- Multidimensional semantic representations emerge from multi-frequency representational similarity learning
- FMCerNet: frequency-aware multi-label learning for fine-grained cervical cytology screening
BibTeX
@article{gallard2026eegbased,
title = {{EEG-based deep learning estimates preterm infants’ cerebral age: Contribution of slow-wave and cross-frequency coupling}},
author = {Alban Gallard and Alexander Billiot and Émilie Bourel-Ponchel and Pauline Brunel and Christèle Chazal and Guy Kongolo and Ghida Ghostine and Fabrice Wallois and Laura Routier and Sahar Moghimi},
journal = {iScience},
year = {2026},
month = oct,
doi = {10.1016/j.isci.2026.118052},
url = {https://www.cell.com/iscience/fulltext/S2589-0042(26)03435-8},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - EEG-based deep learning estimates preterm infants’ cerebral age: Contribution of slow-wave and cross-frequency coupling
AU - Alban Gallard
AU - Alexander Billiot
AU - Émilie Bourel-Ponchel
AU - Pauline Brunel
AU - Christèle Chazal
AU - Guy Kongolo
AU - Ghida Ghostine
AU - Fabrice Wallois
AU - Laura Routier
AU - Sahar Moghimi
PY - 2026
DA - 2026/10/26
JO - iScience
DO - 10.1016/j.isci.2026.118052
UR - https://www.cell.com/iscience/fulltext/S2589-0042(26)03435-8
AB - Reliably estimating cerebral maturation in preterm newborns is crucial for early identification of atypical neurodevelopment. We developed a deep-learning framework using a convolutional neural network (CNN) to predict postmenstrual age (PMA) from spontaneous neural activity in EEG recordings of 165 healthy preterm infants (28–38 weeks PMA). The model achieved a mean absolute error of 0.75 ± 0.60 weeks, with 73.9% of prediction errors within one week. To identify key neurophysiological drivers, we systematically manipulated signal components. Results showed that removing low-frequency activity (<4 Hz) significantly degraded performance and disrupting phase-amplitude coupling selectively impaired prediction accuracy before 34 weeks PMA, while high-frequency activity (>12 Hz) had minimal impact. Our results demonstrate that maturation signatures are primarily driven by slow-wave dynamics and their hierarchical coupling with faster oscillations, which are fundamental hallmarks of early neurodevelopment.
SN - 2589-0042
ER -