Published in IEEE Journal of Biomedical and Health Informatics (under review), 2025
FetalSleepNet transfers adult sleep EEG knowledge to fetal sheep EEG using spectral equalisation and domain adaptation, improving cross-species fetal sleep stage classification.
Published in Neural Networks, 2025
We propose a contrastive graph auto-encoder that learns high-quality node embeddings via joint reconstruction and contrastive objectives, achieving strong performance on graph representation learning benchmarks.
Published in Sleep (under minor revision), 2025
A comprehensive review of fetal sleep physiology, measurement modalities and state classification across humans, sheep and other preclinical models, highlighting implications for neurodevelopment and monitoring.
Published in arXiv preprint, 2024
We study fairness in knowledge tracing for second language acquisition, proposing methods to improve predictive performance while mitigating biases across learner subgroups.
Published in IEEE Sensors 2024, 2024
Using fetal EEG, ECG, EMG and blood pressure signals in a preclinical sheep model, we develop machine learning methods to detect hypoxia–ischemia during labour-like umbilical cord occlusion.
Published in ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2024
We propose GuessKT, a knowledge tracing model that explicitly models students’ guess behaviors to improve the reliability of learning feedback and enhance prediction accuracy across multiple datasets.