Investigating early signs of dementia
As a global health burden, Alzheimer's disease (AD) currently has around 47 million affected individuals worldwide and has global healthcare costs of around !880 billion annually. One of the characteristics of this disease is that it appears in different forms that vary in symptoms as measured by clinicians, and in the patterns of pathology in the brain. This variability makes the disease difficult to manage and clinical trials that probe whether a treatment can slow down, or even cure, AD are time-consuming to plan and execute.
If we could delay the onset of AD by five years, then we could reduce care costs by 36%, saving about !88 billion per year across the EU. To achieve such a goal as delaying the onset of AD, we plan to study how to avoid lifestyle choices and behaviours that could increase AD risk via a better understanding of how the disease develops.
Specifically, this three-year project aims to untangle the variability in AD by analysing relevant data, including from the UK BioBank. We plan to use machine learning algorithms developed by our group to find subtypes of the clinical appearance of AD based on brain imaging data, measurements of cognitive capability, genetics and lifestyle factors. Not only will it shed new light on the nature and variability of disease mechanisms, the study of these AD subtypes is crucial for understanding how genetics and other risk factors influence the patterns of brain shrinkage and loss of brain function years before the disease becomes severe. This discovery will further highlight potential lifestyle interventions at very early stages that may affect or delay disease onset, and will enable enrichment of future clinical trials for specific groups of patients who are likely to benefit from a particular treatment.