Motor Neuron Disease (MND) is a progressive neurodegenerative condition that remains difficult to diagnose early, partly because current biomarkers are limited and disease presentation varies greatly between patients. This project explores how artificial intelligence can support earlier detection and improve understanding of disease mechanisms through the analysis of pathological tissue and digital biomarkers.
The project combines deep learning, explainable AI, and digital pathology to identify patterns associated with abnormal protein aggregation in MND. Working closely with clinical collaborators, the research analyses post-mortem brain and gastrointestinal tissue images to investigate whether minimally invasive biomarkers could support future diagnostic pathways. A key aspect of the work is ensuring that AI models remain biologically interpretable, allowing researchers and clinicians to understand which pathological features are driving model decisions.
Findings so far have shown promising results in distinguishing disease-related tissue patterns and identifying biologically relevant regions associated with protein aggregation, supporting the potential of AI-assisted pathology tools for neurodegenerative disease research.
