Evaluation of novel markers of on-set of PD using machine learning and routine CT brain imaging

This project addresses the priority area of early detection of PD onset to allow timely administration of new treatments.

The main goal in this project is to develop and evaluate an artificial intelligence (AI)-based tool to analyse routine computer tomography (CT) brain scans in hospital libraries to detect PD onset before routine diagnosis. This will lead to a new way to detect people at risk of developing PD and to earlier diagnosis.

PD is a common neurodegenerative disease (around 130, 000 people are affected in the UK, increasing by 2% per year). Currently, it is diagnosed on the basis of examination and movement disorder.

We believe CT brain changes can serve as radiological markers detectable before the onset of movement symptoms. Project work will include developing an AI pipeline, based on deep learning methods, to analyse CT scans to detect the changes and evaluating the changes as markers of PD.

This is an inter-disciplinary project which brings together experts in clinical neuroscience, radiology, electronics, computer science and statistics.