Doctoral research on EEG microstates with variational autoencoders at University College Cork, supervised by Luca Longo . The work asks whether a learned latent geometry produces a microstate alphabet that's more stable across sessions and more behaviourally predictive than classical clustering.

Currently on the desk: screening studies for the registered systematic review of autoencoder-based models for scalp EEG (protocol on OSF Registries, with Morteza Akbari and Luca Longo), and extending the Conv-VaDE architecture search beyond the LEMON dataset.

Recent: the Conv-VaDE paper was presented at XAI 2026 in Fortaleza, Brazil (July 2026), and the full Conv-VaDE study is now published in Brain Informatics (August 2026).

Based in Dublin, Ireland. I travel for research.

open to collaboration seeking funding computing resources

Get in touch →

This is a /now page: a living snapshot of current focus, updated when the focus changes.