A simple brain scan can detect people with early-stage Alzheimer’s disease, a study suggests.
In what could be a breakthrough, researchers have developed an algorithm that can diagnose the condition with up to 98 per cent accuracy.
The computer programme uses standard MRI technology found in most hospitals to produce a result in 12 hours.
Currently, it can take months to diagnose the disease on the NHS and requires a raft of memory and cognitive tests as well as scans.
Researchers from Imperial College London who developed the algorithm, which was tested on more than 400 people, hope it will be rolled out on the NHS by 2025.
The technology works by looking for anomalies in 115 regions of the brain, analysing features such as size, shape and texture. It could help experts work out exactly what causes the memory-robbing condition.
It has previously been used to scan tumours of women with ovarian cancer to judge how the cancer is likely to progress.
Scientists have developed an algorithm that can diagnose the condition with up to 98 per cent accuracy. It uses standard MRI technology found in most hospitals and produces a result in 12 hours (stock image)
What is Alzheimer’s?
Alzheimer’s disease is a progressive, degenerative disease of the brain, in which build-up of abnormal proteins causes nerve cells to die.
This disrupts the transmitters that carry messages, and causes the brain to shrink.
More than 5 million people suffer from the disease in the US, where it is the 6th leading cause of death, and more than 1 million Britons have it.
As brain cells die, the functions they provide are lost.
That includes memory, orientation and the ability to think and reason.
The progress of the disease is slow and gradual.
On average, patients live five to seven years after diagnosis, but some may live for ten to 15 years.
Source: Alzheimer’s Association
Imperial’s Professor Eric Aboagye, who led the study, said: ‘Currently no other simple and widely available methods can predict Alzheimer’s disease with this level of accuracy, so our research is an important step forward.
‘Waiting for a diagnosis can be a horrible experience for patients and their families.
‘If we could cut down the amount of time they have to wait, make diagnosis a simpler process, and reduce some of the uncertainty, that would help a great deal.
‘If all went well, we hope this could be available on the NHS in two to three years.’
Alzheimer’s is the most common form of dementia, affecting over half a million people in the UK and around 6million in the US.
The disease causes brain cells to die and areas of the brain to shrink, which can be seen on a brain scan.
To see how well an MRI scan could pick up early-stage Alzheimer’s, researchers looked at 172 people with the disease who had only mild cognitive impairment.
These minor memory issues, such as forgetting recent events or repeating the same question, are normally precursors for dementia but can be mistaken for the normal memory loss of getting older.
The algorithm accurately identified these people in around three-quarters of cases, even without using memory test results.
It typically takes two to five years for mild cognitive impairment to turn into full dementia and patients with suspected dementia can be monitored for months or even years before being diagnosed.
The algorithm looks for 660 different features in the brain, picking up on very subtle changes that could be missed by the human eye.
Researchers found there are surprising early signs of Alzheimer’s in brain regions never previously linked to the disease, including the cerebellum, which regulates physical activity, and the ventral diencephalon, which is linked to sight and hearing.
The disease is infamous for its effect on memory. But researchers found the hippocampus — the ‘memory centre’ of the brain — picks up only 26 per cent of patients with Alzheimer’s.
Although there is no cure for Alzheimer’s disease, getting a diagnosis quickly helps patients access early support, get treatment to manage their symptoms and plan for the future.
The algorithm developed to analyse brain scans returns a result overnight, currently within 10 to 12 hours.
Professor Aboagye said: ‘This could help people avoid weeks to months of anxiety after first going to see their GP as they wait for results from cognitive tests and to get a diagnosis.’
The Imperial team also looked at scans of people with other neurological conditions, including frontotemporal dementia and Parkinson’s , to see if the algorithm could differentiate between the disorders.
They tested it on 172 people with Alzheimer’s, and 254 people with other neurological conditions or who were healthy.
These included 83 people from a memory clinic, with suspected dementia.
The brain scan performs better, with up to 98 per cent accuracy, than other standard methods used for diagnosis, according to the study authors.
A spinal tap to look at fluid from the brain containing trademark proteins linked to Alzheimer’s disease is 62 per cent accurate.
The brain scan was not compared to PET scans for dementia.