In a groundbreaking study published in 2010, Nature Mental HealthResearchers at Queen Mary, University of London have developed a new method to predict dementia with over 80% accuracy up to nine years before clinical diagnosis. The method, which goes beyond traditional memory tests and measures of brain atrophy, relies on using functional magnetic resonance imaging (fMRI) to detect changes in the brain's default mode network (DMN).
Dementia is a general term for a range of conditions that cause a gradual decline in cognitive function that interferes with daily living and independent activities, affecting memory, thinking, orientation, understanding, calculations, learning ability, language and judgement.
Alzheimer's disease is the most common cause of dementia, accounting for 60-70% of cases. Other types include vascular dementia, Lewy body dementia, and frontotemporal dementia.
Dementia is a progressive disease that worsens over time, often causing significant impairment to daily activities and quality of life. Currently, there is no cure for dementia and treatment focuses primarily on managing symptoms and support for patients and their carers.
Early diagnosis is important because it paves the way for interventions that may slow disease progression, improve quality of life and give patients and their families more time to plan for the future. Traditional diagnostic methods, such as memory tests and brain scans to detect atrophy, often detect the disease only after significant neurological damage has occurred. These methods are not sensitive enough to detect very early changes in brain function that precede clinical symptoms.
“Predicting who will develop dementia in the future is essential to developing treatments that prevent the irreversible loss of brain cells that causes dementia symptoms,” said Charles Marshall, who led the research team from the Centre for Preventive Neurology at the Wolfson Institute for Population Health at Queen Mary University of Science. “Despite improving detection of proteins in the brain that can cause Alzheimer's disease, many people harbor these proteins in their brains and live for decades without ever developing the symptoms of dementia.”
“We hope that our new measurement of brain function will enable us to more accurately determine whether and when someone will actually develop dementia, and therefore identify whether future treatments would be beneficial.”
The study employed a nested case-control study design using data from the UK Biobank, a large biomedical database. The researchers focused on the subset of participants who underwent a functional magnetic resonance imaging (fMRI) scan and were diagnosed with dementia or later developed dementia. The sample consisted of 148 dementia cases and 1,030 controls, matched for age, sex, ethnicity, handedness, and geographic location of the MRI scanning center to ensure a robust comparison group.
Participants underwent resting-state fMRI (rs-fMRI) scans, which measure brain activity by detecting changes in blood flow. The researchers specifically targeted the default mode network (DMN), a network of brain regions that are active at rest and involved in high-level cognitive functions such as social cognition and self-referential thinking.
The researchers analyzed the rs-fMRI data and estimated the effective connectivity between different regions in the DMN using a technique called dynamic causal modeling (DCM). This method goes beyond simple correlations to model the causal influence of one brain region on other brain regions, providing a detailed picture of neural connectivity.
The researchers used these connectivity estimates to train a machine learning model whose goal was to distinguish between people who will develop dementia and those who will not. The training process employed rigorous cross-validation techniques to ensure the model was reliable and to prevent overfitting. Additionally, using similar data and validation techniques, they also developed a prognostic model to predict time to dementia diagnosis.
The predictive model achieved an area under the curve (AUC) of 0.824, demonstrating excellent performance in distinguishing future dementia patients from controls. This level of accuracy is significantly higher than traditional diagnostic methods, which often struggle to detect early-stage dementia.
The model identified 15 key connectivity parameters within the DMN that were significantly different between future dementia cases and controls. Of these, the most significant changes included increased inhibition from the ventromedial prefrontal cortex (vmPFC) to the left parahippocampal gyrus (lPHF) and increased inhibition from the left intraparietal cortex (lIPC) to the lPHF, as well as attenuated inhibition from the right parahippocampal gyrus (rPHF) to the dorsomedial prefrontal cortex (dmPFC).
In addition to diagnostic capabilities, this study also developed a prognostic model to predict time to dementia diagnosis. The model showed a strong correlation (Spearman's ρ = 0.53) between predicted and actual time to diagnosis, indicating its potential to provide a valuable timeline of disease progression. The predictive power of these connectivity patterns suggests that DMN alterations may serve as early biomarkers of dementia, providing insight into the course of the disease years before clinical symptoms appear.
Furthermore, this study investigated the relationship between changes in DMN connectivity and various risk factors for dementia. A significant association was found between social isolation and DMN dysfunction, suggesting that social isolation may exacerbate dementia-related neural changes. This finding highlights the importance of considering environmental and lifestyle factors in dementia risk and paves the way for potential avenues of intervention.
“By using these analytical techniques on large datasets, we can not only identify people at high risk of dementia, but also learn what environmental risk factors pushed these people into the high-risk zone,” said co-author Samuel Herreira. “There is great potential to apply these methods to different brain networks and populations to help us better understand the interplay between environment, neurobiology and disease in both dementia and possibly other neurodegenerative diseases. fMRI is a non-invasive medical imaging tool, taking about six minutes in an MRI scanner to collect the necessary data, so it can be integrated into existing diagnostic pathways, especially where MRI is already used.”
Despite the promising results, there are some caveats to consider. One limitation of this study is that it uses data from the UK Biobank, which is not fully representative of the general population. Participants in this cohort tended to be healthier and less socio-economically disadvantaged. Future studies should validate these findings in more diverse and representative samples.
“One in three people with dementia do not have a formal diagnosis, so there is an urgent need to improve how we diagnose people with dementia. This will become even more important as dementia becomes a treatable disease,” Julia Dudley, head of strategic research programmes at Alzheimer's Research UK, told the Science Media Centre.
“This study provides intriguing insights into early signs that people may be at higher risk of developing dementia. This technology needs to be validated in further studies, but if so, it could be a promising addition to the toolkit of ways to detect dementia-causing diseases as early as possible. Early and accurate diagnosis allows for personalized care and support, and is key to accessing the first cures in the near future.”
Eugene Duff, senior research fellow at the UK Dementia Research Institute at Imperial College London, added: “This study shows that advanced analysis of brain activity measured using MRI can predict a future dementia diagnosis. Early diagnosis of dementia is valuable for a number of reasons, particularly as improved pharmaceutical treatments become available.”
“Measurements of brain activity may complement cognitive, blood and other markers for identifying people at risk of dementia. The brain modelling approach they use has the advantage that it may reveal brain processes affected at an early stage of the disease. However, the study cohort of diagnosed patients was relatively small (103 cases). Further validation and direct comparison of predictive markers is needed.”
The study, “Early Detection of Dementia Through Effective Connectivity of the Default Mode Network,” was authored by Sam Ereira, Sheena Waters, Adeel Razi, and Charles R. Marshall.
