Predicting dementia nine years before symptoms

Researchers at Queen Mary’s Centre of Preventive Neurology have developed a new method for predicting dementia with over 80% accuracy and up to nine years before a diagnosis. The new method provides a more accurate way to predict dementia than memory tests or measurements of brain shrinkage, two commonly used methods for diagnosing dementia.

Predicting who is going to get dementia in the future will be vital for developing treatments that can prevent the irreversible loss of brain cells that causes the symptoms of dementia.

The team, led by Professor Charles Marshall, developed the predictive test by analysing functional MRI (fMRI) scans to detect changes in the brain’s ‘default mode network’ (DMN). The DMN connects regions of the brain to perform specific cognitive functions and is the first neural network to be affected by Alzheimer’s disease.

The researchers used fMRI scans from over 1,100 volunteers from UK Biobank, a large-scale biomedical database and research resource containing genetic and health information from half a million UK participants, to estimate the effective connectivity between ten regions of the brain that constitute the default mode network.

The researchers assigned each patient with a probability of dementia value based on the extent to which their effective connectivity pattern conforms to a pattern that indicates dementia or a control-like pattern.

They compared these predictions to the medical data of each patient, on record with the UK Biobank. The findings showed that the model had accurately predicted onset of dementia up to nine years before an official diagnosis was made, and with greater than 80% accuracy. In the cases where the volunteers had gone on to develop dementia, it was also found that the model could predict within a two-year margin of error exactly how long it would take that diagnosis to be made.

Predicting which patients may later have dementia could allow for better design, control and management of trials – as you’d only be including people who are actually likely to get dementia. This means life-saving treatments can be developed faster.

We hope one day these scans will become standard practice, allowing doctors to diagnose patients early on and that collectively they’ll provide the data to develop strategies to prevent dementia in the first place.

We’re looking for a partner to help develop this technology further and bring it to patients.


Dr Monika Hamilton –


Professor Charles Marshall 

Professor and Honorary Consultant Neurologist

More from our pipeline

Enhancing  immunotherapy in lung cancer

Restorative treatments for osteoarthritis

Repurposing an anti-cancer agent for heart failure