A new machine learning powered brain scan can detect when a person has early stage Alzheimer’s with near perfect accuracy, and researchers are hopeful it will become available for use in the coming years.
Researchers at the Imperial College London, in the UK, used an algorithm that already existed to detect cancerous tumors, and repurposed it to detect early Alzheimer’s with 98 percent accuracy.
Detecting Alzheimer’s early allows for treatment to start earlier, gives doctors more time to plan how to approach treatment, and even allows the patient themselves to plan for the future while they are still in a sound mind.
While the machine learning system is still in its early stages, the researchers are hopeful it will be available for consumers within the next few years.
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)
‘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,’ Professor Eric Aboagye, who led the research, said in a statement.
‘Many patients who present with Alzheimer’s at memory clinics do also have other neurological conditions, but even within this group our system could pick out those patients who had Alzheimer’s from those who did not.’
POTENTIAL CURES AND TREATMENTS FOR ALZHEIMER’S
Stem cells are ‘building block’ cells that can develop into many different cell types, including brain or nerve cells.
They have the potential to repair brain damage caused by neurological conditions, such as dementia.
Vaccines and antibodies
Immunotherapy involves boosting the body’s own defences to fight disease.
It is used to treat cancer and Covid.
Some studies are looking at vaccines that block a build-up of proteins in the brains of dementia patients.
Other studies have used monoclonal antibodies – which involves injecting molecules that have been primed in a lab and designed to target plaque in the brain.
The UK CLARITY study is measuring how effective the monoclonal antibody BAN2401 (lecanemab) is at preventing or delaying the very early stages of Alzheimer’s disease.
Another therapy, called aducanumab, is currently under consideration by the US Food and Drug Administration (FDA).
Developing new medicines to treat dementia takes many years and millions of pounds.
Repurposing existing drugs used for other conditions is another, often quicker, way of finding medicines to treat dementia.
Current medicines being explored as possible treatments for Alzheimer’s and vascular dementia include those used for:
- type 2 diabetes
- high blood pressure
- rheumatoid arthritis
Researchers, who published their findings in Communications Medicine this week, gathered data from 400 patients who already have been diagnosed with either early or late stage Alzheimer’s.
They used a machine learning system used to classify cancerous tumors, and instead used it on the brain.
The brain was split into 115 regions, and more than 600 features to be scanned, like the size, shape and texture of each section.
Using an MRI machine, which a majority of hospitals in developed countries already have, doctors could get a scan of a person’s brain that the machine learning system could read.
The system correctly identified a case of Alzheimer’s in 98 percent of participant’s brain scans.
In nearly 80 percent of the scans, it was able to tell whether the person had late or early stage Alzheimer’s.
‘Although neuroradiologists already interpret MRI scans to help diagnose Alzheimer’s, there are likely to be features of the scans that aren’t visible, even to specialists,’ Dr Paresh Malhotra, a neurologist at the college, said.
‘Using an algorithm able to select texture and subtle structural features in the brain that are affected by Alzheimer’s could really enhance the information we can gain from standard imaging techniques.’
There are currently no cures for Alzheimer’s and treatments are more focused on managing symptoms then stopping the progression of the disease.
Only one drug has ever been approved by regulators to stop cognitive decline associated with Alzheimer’s – Biogen’s Aduhelm. The drug’s use is extremely limited as many experts doubt that it works.
Finding Alzheimer’s early is still valuable, though. A person who knows that they may soon begin losing their memory can better prepare for a life where they are no longer all there.
This includes setting up a care situation – whether a family member or a private nurse of some sort.
It also allows the person to arrange post-life affairs in order while they still have not experienced any significant memory loss.
Researchers are also hopeful that their machine learning system can help in future drug trials, as it could identify early-stage Alzheimer’s patients before other methods can, and use then as participants in trials.
‘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,’ Aboagye said.
‘Our new approach could also identify early-stage patients for clinical trials of new drug treatments or lifestyle changes, which is currently very hard to do.’
More than six million Americans are living with Alzheimer’s, according to the Alzheimer’s Association, and it it one of the leading killers of people over the age of 65.