AI can accurately predict who will develop dementia

Pioneering technology using artificial intelligence in the UK and US can accurately predict which people will within two years develop dementia.

A study forecast which patients would develop dementia with up to 92 per cent accuracy, surpassing other research methods. University of Exeter researchers used data from 15,300 patients in the US and found that a form of Artificial Intelligence (AI) called machine learning can help to determine dementia sufferers.

The technique works by spotting hidden patterns in the data and learning who is most vulnerable. The researchers say the algorithm could help to reduce the number of people wrongly diagnosed with dementia.

Prof David Llewellyn, an Alan Turing fellow at the University of Exeter, oversaw the study. “We’re now able to teach computers to accurately predict who will go on to develop dementia within two years," he said.

“We’re also excited to learn that our machine-learning approach was able to identify patients who may have been misdiagnosed.

“This has the potential to reduce the guesswork in clinical practice and significantly improve the diagnostic pathway, helping families access the support they need as swiftly and as accurately as possible.”

From 2005 and 2015, one in 10 attendees (1,568) at the memory clinic received a new diagnosis of dementia within two years of their visit.

The researchers found for the first time that about 8 per cent (130) of the diagnoses appeared to be made in error, in which case they were reversed.

The study, published in JAMA Network Open and funded by Alzheimer’s Research UK, showed that machine-learning models accurately identified more than 80 per cent of those diagnoses.

The research suggests AI can not only accurately pinpoint who will be diagnosed with dementia, but has the potential to improve the accuracy of the diagnoses.

Machine learning works by using patient information routinely available in the clinic, such as memory and brain function, performance on cognitive tests and lifestyle factors.

The team plans follow-up studies to evaluate the practical use of the machine-learning method in clinics, to assess whether it can be used widely to improve diagnosis, treatment and care.

The researchers analysed data from people who attended a network of 30 National Alzheimer’s Co-ordinating Centre memory clinics in the US.

The attendees did not have dementia at the start of the study, though many were experiencing problems with memory or other brain functions.

“Artificial intelligence has huge potential for improving early detection of the diseases that cause dementia and could revolutionise the diagnosis process for people concerned about themselves or a loved one showing symptoms," said Dr Rosa Sancho, head of research at Alzheimer’s Research UK.

“This technique is a significant improvement over existing alternative approaches and could give doctors a basis for recommending lifestyle changes and identifying people who might benefit from support or in-depth assessments.”