Research

Graph learning for networks and health

Machine learning methods for understanding complex networks and addressing challenges in healthcare and biomedical data analysis. Doctoral research (2020 to 2025): social network analysis and drug discovery with graph convolutional networks.

Graph neural networks and representation learning

Graph convolutional networks and representation-learning methods that treat relationships as first-class data, learning from structure and node features to find influential nodes and characterise complex graphs.

Graph convolutional networks
Representation learning
Influential node identification
Structural and feature-based learning

Social network analysis

The structure, organisation and dynamics of online social networks: influential users, community structure, and the overlap between communities.

Influential users and nodes
Community detection
Overlapping communities
Network topology

Drug discovery and biomedical applications

How graph-based representations can support computational drug discovery, where molecular and biomedical entities are understood through their relationships. This is an exploratory research direction, not a claim of clinical or therapeutic outcome.

Graph-based biomedical modelling
Machine learning for drug discovery
Biomedical relationship learning

Healthcare analytics

Machine learning and deep learning for healthcare prediction, including early cardiovascular disease detection and a submitted study on coronary artery disease prediction from MRI images. These are research investigations, not clinical deployments.

Healthcare data analysis
Disease prediction
Medical image analysis

Open to research collaboration

For research discussions, collaboration, student enquiries and correspondence.

Email All contact details