Scientists have developed an AI tool which is more than 80 per cent accurate at predicting whether people with early signs of dementia will remain stable or develop Alzheimer’s disease, a study has found.
The team from Cambridge University say the technology, called Brain Health, is three times more accurate than standard “markers” for the disease such as shrinking grey matter in the brain, declining scores for ‘cognitive’ tests, or a clinical diagnosis.
Researchers say this new method – which is based on cognitive tests and an MRI scan – could reduce the need for invasive and costly diagnostic tests.
It is also thought that the tool could increase the speed and accuracy of diagnosis, and this may lead to treatments such as lifestyle changes and new drugs being more effective because of the disease being at an earlier stage.
If a further clinical trial is successful, the tool could be available on the health service in as little as two to three years, researchers told i.
Professor Zoe Kourtzi, of Cambridge University, who helped develop the technology, said: “We’ve created a tool which, despite using only data from cognitive tests and MRI scans, is much more sensitive than current approaches at predicting whether someone will progress from mild symptoms to Alzheimer’s – and if so, whether this progress will be fast or slow.
“This has the potential to significantly improve patient wellbeing, showing us which people need closest care, while removing the anxiety for those patients we predict will remain stable.”
He added that the tool will “help remove the need for unnecessary invasive and costly diagnostic tests” at a “time of intense pressure on healthcare resources”.
The main cause of dementia is Alzheimer’s disease, and it is believed to accounts for up to 80 per cent of cases.
Early detection is crucial as this is when treatments are likely to be most effective, yet early dementia diagnosis and prognosis may not be accurate without invasive or expensive tests such as positron emission tomography (PET) scans or lumbar puncture, which are not available in all memory clinics.
As a result, up to a third of patients may be misdiagnosed and others diagnosed too late for treatment to be effective.
10 early signs of dementia
- Memory loss that disrupts daily life
- Challenges in planning or solving problems
- Difficulty completing familiar tasks
- Confusion with time or place
- Trouble understanding visual images and spatial relationships
- New problems with words in speaking or writing
- Misplacing things and losing the ability to retrace steps
- Decreased or poor judgment
- Withdrawal from work or social activities
- Changes in mood and personality
The AI test is being developed at the same time as scientists work on diagnostic blood tests for Alzheimer’s that they hope could be available within five years.
It is not clear whether the two types of tests could be used in conjunction or for separate purposes – assuming further tests confirm the benefits of each and they become available on the health care system.
Timothy Rittman, of Cambridge University, told i: “Ideally when a person is referred to a memory clinic they would have both a brain scan and a blood test.
“Blood tests are only specific for one type of dementia – Alzheimer’s disease. We can extend the machine learning algorithm to recognise different types of dementia. Blood tests may be diagnostic but they are not prognostic [predictive]. Our AI tool can predict future decline.”
Ivan Koychev, consultant neuropsychiatrist at Oxford University Hospitals NHS Foundation Trust, who was not involved in the research, described the AI tool as “significant” adding “it is entirely possible that the two methods – AI and blood tests – will be better suited for different patients”.
The researchers now hope to extend their model to other forms of dementia, such as vascular dementia and frontotemporal dementia, using different types of data, such as markers from blood tests.
The scientists developed a machine learning model able to predict whether and how fast an individual with mild memory and thinking problems will progress to developing Alzheimer’s.
To build their model, the researchers used routinely collected, non-invasive, and low-cost patient data – cognitive tests and structural MRI scans showing grey matter atrophy – from more than 400 people who were part of a research group in the US.
They then tested the model using real-world patient data from a further 600 participants from the US cohort and – importantly – longitudinal data from 900 people from memory clinics in the UK and Singapore.
The AI tool was able to distinguish between people with stable mild cognitive impairment and those who progressed to Alzheimer’s within a three-year period.
It correctly identified people who went on to develop the disease in 82 per cent of cases, and correctly identify those who didn’t in 81 per cent of cases.
While the researchers tested the algorithm on data from a research cohort, it was validated using independent data that included almost 900 people who attended memory clinics in the UK and Singapore.
The researchers say this shows it should be applicable in a real-world patient clinical setting.
Dr Richard Oakley, associate director of research and innovation at the Alzheimer’s Society, said the research “brings further hope for improving the speed and accuracy of dementia diagnosis, which currently far too many are missing out on”
“With further validation, this model has the potential to significantly reduce cases of people with dementia from being undiagnosed or misdiagnosed.”
Dr Susan Kohlhaas, executive director of research and partnerships at Alzheimer’s Research UK, which co-funded the study, said: “This study shows that AI holds potential in improving the way we detect dementia early and more accurately and forms an important part of our work on early detection.
“It will be particularly interesting to see how this technology could be combined with other promising new developments in Alzheimer’s diagnosis, such as blood tests.”
The study, published in the journal eClinical Medicine was funded by Wellcome, the Royal Society, Alzheimer’s Research UK, the Alzheimer’s Drug Discovery Foundation Diagnostics Accelerator, the Alan Turing Institute, and the National Institute for Health and Care Research Cambridge Biomedical Research Centre.