Neonatal Hypoxic-ischaemic Encephalopathy Classification from the EEG and HRV Signals Using a Conformer based Masked Autoencoder

2026-07-26

AI Summary

This study addresses the challenge of classifying the severity of neonatal hypoxic-ischemic encephalopathy (HIE) by proposing MAEConformer, a novel framework that integrates the Conformer architecture with a masked autoencoder for self-supervised representation learning from multimodal physiological signals—specifically electroencephalography (EEG) and heart rate variability (HRV). To enhance time-frequency reconstruction fidelity, the method introduces a multi-resolution short-time Fourier transform loss and leverages large-scale unlabeled data for pretraining before transferring to downstream classification tasks. Experimental results demonstrate that the proposed approach achieves state-of-the-art performance, yielding AUC scores of 97.19% and 96.56% on binary and four-class HIE classification tasks using EEG, respectively, and 82.42% AUC with HRV, significantly outperforming existing supervised and self-supervised methods.

Abstract

In this paper, we propose the MAEConformer, a novel self-supervised learning framework that combines the Conformer architecture with the Masked Autoencoder (MAE) paradigm for large-scale representation learning from unlabelled electroencephalography (EEG) and heart rate variability (HRV) signals. By integrating convolutional operations with Transformer-based self-attention, MAEConformer effectively captures both local temporal patterns and long-range contextual dependencies in physiological time series. To enhance reconstruction fidelity and representation quality, a multi-resolution short-time Fourier transform (MR-STFT) loss is incorporated alongside the reconstruction objective, enabling the model to jointly learn temporal and spectral characteristics across multiple scales. Modality-specific EEG and HRV MAEConformer models were pretrained on 6,030h and 4,868h of unlabelled recordings, respectively, and subsequently transferred to expert-annotated downstream tasks. Experimental results demonstrate that the learned representations provide strong transferability and data efficiency. In EEG-based hypoxic ischemic encephalopathy (HIE) severity classification, the pretrained MAE-EEG model achieved test AUCs of 97.19% and 96.56% for binary and four-class classification tasks, respectively, outperforming a range of state-of-the-art supervised and self-supervised baselines. On the HRV-based HIE severity classification task, MAE-HRV achieved a test AUC of 82.42%, surpassing both self-supervised Transformer-based and supervised convolutional baselines. These findings demonstrate the effectiveness of MAEConformer for learning robust and transferable representations across multiple physiological modalities.

Problem

• Neonatal Hypoxic-ischaemic Encephalopathy
• EEG
• HRV
• Severity Classification

Innovation

• Conformer
• Masked Autoencoder
• Self-supervised Learning
• Multi-resolution STFT Loss
• Physiological Signal Representation