Grant: CIHR Fellowship

By 2031, there will be a projected 674,000 Canadians living with dementia. Early detection of dementia risk will thus be critical to reducing dementia prevalence. Current evidence suggests that there are a number of modifiable risk factors which may lower dementia risk by as much as 40%. These risk factors include: educational attainment, alcohol intake, obesity, smoking, and physical inactivity, among others. Importantly, there are now opportunities to use the power of artificial intelligence, specifically machine learning (ML), to enhance our ability to detect early risk for dementia based on the presence (or lack thereof) of these risk factors. However, no ML algorithm has yet been developed to 1) identify brain neuroimaging markers which predict cognitive performance; or 2) predict neuroimaing markers of cognitive performance from modifiable risk factors. Thus, I will use data from the UK Biobank study (N~45,000) to develop ML algorithms which will identify brain imaging markers of cognitive performance that can be linked to modifiable risk factors. The results of my study could markedly improve population health by helping to facilitate targeted interventions to prevent cognitive decline through early detection of modifiable risk from neuroimaging.