New AI technology in development to predict dementia risk from eye test

Announced today, the technology is the focus of work by the NeurEYE research team, led by the University of Edinburgh, with Glasgow Caledonian University.

In an effort to create a large-scale dataset optimal for AI analysis, the research team have already collected almost a million anonymised eye scans from a collaboration with opticians across Scotland.

Using AI and machine learning, they will be able to analyse the image data, linked to other relevant data – such as demographics, pre-existing conditions and treatment history.  The technology will allow researchers to identify patterns that could indicate a person’s risk of developing dementia, as well as determine a broader picture of their brain health.

Eye health is an expanding area of dementia research.  Scientists have recently identified vision loss as a modifiable risk factor for dementia.  The NeurEYE research project not only lays the foundation for future advances in dementia research, determined through the eye. But it could also allow the earlier detection of Alzheimer’s disease and other dementias – leading to faster development of treatments, more timely diagnosis, and better monitoring of patients suffering from these debilitating diseases.

Collating a dataset of such scale will also improve the quality of the algorithms used to predict dementia.  The large data set helps provide a much more representative sample of the population on which AI algorithms can be trained – reducing bias and producing more equitable findings, which better serve our whole communities.

Approval comes from the Public Benefit and Privacy Panel for Health and Social Care, a part of NHS Scotland and the data will be held safely in the Scottish National Safe Haven, commissioned by Public Health Scotland, which provides a secure platform for the research use of NHS electronic data.

NeurEYE is the second project funded and supported by NEURii, a landmark global collaboration between the pharmaceutical company Eisai, Gates Ventures, the University of Edinburgh, the medical research charity LifeArc, and Health Data Research UK (HDR UK), the national institute for health data science.  The collaboration supports scientists, clinicians and industry to work together on innovative digital projects, aimed at developing real world solutions to tackling neurodegenerative conditions such as Alzheimer’s disease.

Professor Andrew Morris, Director of HDR UK, said:

“The ability to identify dementia risk via routine eye tests would be a truly significant step forward, in our efforts to combat Alzheimer’s disease – already a significant public health issue, and predicted to affect 1.6 million people in the UK by 2050.

 

“The more we can learn from large-scale, targeted, data-based research like the NeurEYE project, the better we can understand how Alzheimer’s and similar diseases work in the brain itself – and the faster we will discover how to stop it in its tracks, or prevent it from taking hold at all.

 

“The NEURii research collaboration is an excellent example of data scientists, industry, and healthcare professionals working together to unlock the power of health data for public benefit.  Brain health diseases cause untold suffering worldwide, for millions of people and their families.  At HDR UK, working in partnership with the UK Dementia Research Institute and international partners, we are delighted to be a supportive partner in NEURii and other research initiatives in brain health, which are showing enormous promise in tackling one of the major health challenges of the 21^st century.”

The NEURii initiative is part of HDR UK’s Brain Health partnership programme – a host of projects where HDR UK is working with partners to accelerate work into neurodegenerative diseases through the trustworthy use of large-scale data.  Projects such as the Dementia Trials Accelerator (DTA), for example, will expand the opportunities for those at risk of, or in early stages of dementia, to take part in clinical trials.  Several other partner collaborations, such as the Motor Neurone Disease (MND) Research Catalyst, the Alzheimer’s Disease (AD) Data Initiative Data Pilots, and the Traumatic Brain Injury (TBI) Reporter, are all highly active projects aimed at bringing data science solutions to the global fight against neurodegenerative disease.