Research

Multimodal neuroimaging for schizophrenia (current)

My postdoctoral work at the Complex Networks & Brain Dynamics Group, ICS CAS Prague focuses on building computational models that characterise and predict clinical outcomes in schizophrenia. I work with structural and functional MRI, combining graph-based representations with deep learning to capture inter-individual variability across multi-site datasets. A central challenge is generalising models trained on one clinical site to others — which motivates my work on domain adaptation and semi-supervised learning for neuroimaging.

Current projects:

EEG and graph neural networks for Alzheimer’s disease (PhD)

During my PhD at Coventry University and A*STAR (Singapore), I developed graph neural network models for EEG-based characterisation of Alzheimer’s disease. This included designing novel functional connectivity measures using bispectral analysis, building explainable GNN classifiers with adaptive gating mechanisms, and conducting a comprehensive survey of GNN methods for EEG classification. A recurring theme was making models interpretable — understanding not just whether a classifier works, but which brain regions and frequency interactions it relies on.

Education

PhD, Mathematical and Statistical Modelling (Cotutelle)
Coventry University & A*STAR Institute for Infocomm Research, 2020–2024
Network Inference and Graph Learning in Characterising Alzheimer’s Disease
Supervised by Dr Fei He (Coventry) and Dr Min Wu (A*STAR)

MSc, Data Science and Computational Intelligence
Coventry University, 2019–2020

BSc, Cognitive Science
Aarhus University, 2016–2019