One of the key features of dementia is the buildup of a toxic protein called amyloid in the walls of arteries in the brain. This accumulation happens because the normal clearance of waste along tiny channels in blood vessel walls, a process known as Intramural Periarterial Drainage (IPAD), begins to fail. Critically, amyloid can build up silently for 10–20 years before any symptoms appear, meaning current treatments are often given far too late. What remains unknown is how many arteries must become clogged with amyloid before symptoms start to develop. 

This project aims to develop an AIdriven understanding of IPAD using computational algorithmic tools to predict the point at which these pathways become blocked beyond repair and the disease process becomes irreversible. Analysing complex data from brain scans and postmortem tissue presents a significant technical challenge, making this an ideal cross-disciplinary studentship. The findings will deepen our understanding of the long development of dementia and help pinpoint the crucial window of opportunity for effective treatments. 

Supervisors: Roxana Carare, Professor of Neuroscience; and Dr Zehor Belkhatir Lecturer in the School of Electronics and Computer Science.