Develop an innovative methodology to predict future cognitive decline
Test the methodology in ongoing trials for dementia prevention
Develop solutions to address the ethical, legal and social issues surrounding computer-derived prediction of dementia risk.
Alzheimer’s disease is a progressive degenerative disease that results in a loss of brain cells. It is the most common form of dementia in the elderly.
There is now compelling evidence that Alzheimer’s disease takes hold in the brain decades before dementia symptoms appear, providing a window of opportunity for preventative intervention.
In addition, research has identified several risk factors associated with dementia. While there are some risk factors you can’t control, such as genetics or age, some risk factors can be managed through lifestyle changes such as smoking, drinking, physical activity levels and diet.
A multidomain lifestyle intervention model tested by consortium members showed beneficial effects on cognition among older adults at risk for dementia from the general population. This pioneer FINGER (Finnish Geriatric Intervention Study to Prevent Cognitive Impairment and Disability) randomised controlled trial demonstrated that a two-year multimodal lifestyle intervention consisting of nutritional guidance, exercise, cognitive training, social stimulation, and control of vascular risk factors shows an improvement in cognition and other related outcomes in older people at increased risk of dementia.
However, this intervention effectiveness may be depending on a personalised approach methodology that does not exist including accurately identifying at-risk individuals who are most likely to benefit.
While there is a large body on literature on personalised medicine studies focusing on predicting the transition of mild cognitive impairment to Alzheimer’s disease, there exists a clear gap of methodology for:
The Pattern-Cog project will address these gaps by developing and validating methods to predict future cognitive decline based on clinical data from multiple sources. Our focus will be in the pre-symptomatic phase of dementia and on persons at-risk.
Instead of a standard machine learning approach, we propose an innovative concept of personalised ageing pattern rooted in data from healthy individuals.
We will develop, assess and validate a new methodology to evaluate future cognitive decline by integrating routine data from multiple data resources (i.e. brain scans, neuropsychological test scores, behavioral data and information about risk and lifestyle factors) from:
The methodology will then be tested in ongoing trials for dementia prevention (multidomain lifestyle intervention with and without putative disease-modifying drugs).
By combining cohorts and resources, we rely on enabling big data. This entails that research data management will play a central role in this project.
