Research

My research is centred on the design and evaluation of artificial-intelligence systems that need to be understandable, robust and useful in applied settings.

Explainable Artificial Intelligence

I work on methods for evaluating and improving local explanations, with particular interest in neighbourhood-based explanation methods, sufficiency and necessity criteria, and the relationship between an explanandum and its local context.

Generative AI and synthetic biomedical data

A second strand of my research examines the quality, diversity and reliability of generative models for biomedical imaging. This includes model-collapse detection, image normalisation, synthetic-data quality assessment and evaluation strategies for GAN-generated medical imagery.

Healthcare AI, fairness and decision support

I am interested in the practical use of machine learning in healthcare, including biomedical image analysis, physiological-signal analysis, clinical decision support, bias detection and fairness evaluation.

Human-centred and organisational AI

My broader work includes affective computing, human-centred evaluation, organisational analytics and data-driven decision support. Across these areas, the common theme is the rigorous evaluation of AI systems in contexts where model behaviour needs to be interpretable and defensible.

Research affiliations and service

I am involved in national research networks including ADAPT and Lero and contribute to programme committees and conference organisation in AI, explainable AI, machine vision and signal processing.