Artificial Intelligence (AI) can predict who will develop dementia within two years with 92 per cent accuracy, a new study has revealed.
And as well as its high degree of success in accurately telling who will go on to develop dementia, the algorithm could also help reduce the number of people who may have been wrongly diagnosed.
The study, using data from more than 15,300 patients in the United States, reveals the role of machine learning in spotting hidden patterns in data to learn who is most at risk.
Researchers from the University of Essex analysed data from people who attended a network of 30 National Alzheimer’s Coordinating Center 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.
In the study timeframe between 2005 and 2015, one in ten attendees (1,568) received a new diagnosis of dementia within two years of visiting the memory clinic.
The research found that the machine learning model could predict these new dementia cases with up to 92 per cent accuracy – and far more accurately than two existing alternative research methods.
The researchers also found for the first time that around eight per cent (130) of the dementia diagnoses appeared to be made in error, as their diagnosis was subsequently reversed.
Machine learning models accurately identified more than 80 per cent of these inconsistent diagnoses. AI can not only accurately predict who will be diagnosed with dementia, found the study – which was funded by Alzheimer’s Research UK – it also has the potential to improve the accuracy of these diagnoses.
Professor David Llewellyn, an Alan Turing Fellow based at the University of Exeter, who oversaw the study, said: “We’re now able to teach computers to accurately predict who will go on to develop dementia within two years.
“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.”
Dr Rosa Sancho, head of research at Alzheimer’s Research UK, said: “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.
“This technique is a significant improvement over existing alternative approaches and could give doctors a basis for recommending life-style changes and identifying people who might benefit from support or in-depth assessments.”
Dr Janice Ranson, Research Fellow at the University of Exeter added “We know that dementia is a highly feared condition. Embedding machine learning in memory clinics could help ensure diagnosis is far more accurate, reducing the unnecessary distress that a wrong diagnosis could cause.”
The researchers found that machine learning works efficiently, using patient information routinely available in clinic, such as memory and brain function, performance on cognitive tests and specific lifestyle factors.
The team now plans to conduct follow-up studies to evaluate the practical use of the machine learning method in clinics, to assess whether it can be rolled out to improve dementia diagnosis, treatment and care.
Young-onset Alzheimer’s prevention trial to launch
The Primary Prevention Trial will target participants up to 25 years before the expected onset of dementia
An international clinical trial is being launched aimed at preventing young-onset Alzheimer’s disease in people genetically destined to develop the illness.
Unlike most other Alzheimer’s prevention trials, this will enrol people before the disease has taken hold, up to 25 years before the expected onset of dementia.
The Primary Prevention Trial, led by Washington University School of Medicine in St. Louis, will investigate whether gantenerumab — an investigational antibody under development for Alzheimer’s disease by Roche and Genentech — can clear a key Alzheimer’s protein called amyloid beta, and slow or stop the disease.
Amyloid is the chief component of plaques that dot the brains of people with the disease.
Many scientists suspect the disease originates from the buildup of amyloid plaques in the brain that start to develop up to two decades before symptoms of dementia begin.
“Overwhelming evidence suggests that the most effective way to slow or stop amyloid beta is to prevent it from building up in the first place, but most of the drugs targeted to this protein have been tested in people who already have at least some early signs of the disease, such as memory loss – when the disease is far enough along that reducing amyloid alone isn’t likely to stop it,” said Dr Eric McDade, an associate professor of neurology and the trial’s principal investigator.
“We’ll be recruiting participants as young as 18. In many ways, this trial will be a necessary test of the amyloid hypothesis, which has had a major influence on Alzheimer’s research and drug development over the past 30 years.”
The new trial involves families with rare genetic mutations that cause Alzheimer’s at a young age – typically in a person’s 50s, 40s or even 30s.
A parent with such a mutation has a 50 per cent chance of passing the genetic mutation to a child, and any child who inherits the mutation is all but guaranteed to develop symptoms of dementia near the same age as his or her parent.
This certainty gives researchers an opportunity to evaluate the effectiveness of drugs designed to prevent Alzheimer’s.
More than $130million has been earmarked for the trial, including grants totalling an estimated $97.4million from the National Institute on Aging (NIA) of the National Institutes of Health (NIH), $14million from the Alzheimer’s Association and the GHR Foundation, and up to $11.5million from longtime Washington University benefactor Joanne Knight of St. Louis and family, who have long supported Alzheimer’s research at Washington University. In addition, the university has pledged to raise an additional $6.5 million.
The trial is being conducted in close partnership with Roche and Genentech, which also is providing significant funding.
“We are thrilled to be part of this important clinical trial in one of the earliest stages of Alzheimer’s studied to date,” said Dr Rachelle Doody, global head of neurodegeneration at Roche and Genentech.
“Our vision has always been to detect Alzheimer’s early, before damage in the brain is irreversible, offering tools and treatment all along the journey for people at risk of the disease.
“Close collaboration between industry, academia and patients is so critical to achieve this and tackle the complex challenge of this disease.”
The trial will recruit people with rare, early-onset forms of the disease, but the results also will further understanding of Alzheimer’s overall, which could benefit the millions of people living with the more common form, which affects people later in life.
Dr McDade and his team are studying about 230 participants from families that carry genetic mutations that lead to early-onset Alzheimer’s disease.
The participants come from sites on five continents and have no or very few amyloid deposits.
The trial will test gantenerumab over four years, with a goal of determining whether early treatment will prevent the buildup of the toxic protein.
COVID ‘can cause more signs of brain damage than Alzheimer’s’
New research found higher levels of seven markers of neurodegeneration in hospitalised COVID patients
Patients hospitalised for COVID-19 had higher levels over the short term of blood proteins known to rise with neurological damage than non–COVID-19 patients diagnosed with Alzheimer’s disease, a new study has found.
The research – conducted over two months early in the COVID-19 pandemic – found higher levels of seven markers of brain damage, or neurodegeneration, in patients with COVID-19 with neurological symptoms than those without them, and much higher levels in patients that died in the hospital than in those discharged and sent home.
A second analysis by the team from NYU Grossman School of Medicine found that a subset of the damage markers in patients hospitalised with COVID-19, over the short term, were significantly higher than in patients diagnosed with Alzheimer’s disease, and in one case more than twice as high.
“Our findings suggest that patients hospitalised for COVID-19, and especially in those experiencing neurological symptoms during their acute infection, may have levels of brain injury markers that are as high as, or higher than, those seen in people who have Alzheimer’s disease,” says lead author Dr Jennifer A. Frontera, professor in the Department of Neurology at NYU Grossman School of Medicine.
“Traumatic brain injury, which is also associated with increases in these biomarkers, does not mean that a patient will develop Alzheimer’s disease or related dementias later on, but does increase the risk of it,” says senior author Dr Thomas M. Wisniewski, the Gerald J. and Dorothy R. Friedman Professor in the Department of Neurology and director of the Center for Cognitive Neurology at NYU Langone.
“Whether that kind of relationship exists in those who survive severe COVID-19 is a question we urgently need to answer with ongoing monitoring of these patients.”
The study, conducted from March to May 2020, identified 251 people who, although 71 years on age on average, had no record or symptoms of cognitive decline or dementia before being hospitalised for COVID-19.
These patients were then divided into groups with and without neurological symptoms during their acute COVID-19 infection, when patients either recovered and were discharged, or died.
The research team also, where possible, compared marker levels in the COVID-19 group to patients in the clinical core cohort of NYU Langone’s Alzheimer’s Disease Research Center, an ongoing, long-term study at NYU Langone Health.
None of these 161 control patients (54 cognitively normal, 54 with mild cognitive impairment, and 53 diagnosed with Alzheimer’s disease) had COVID-19. Brain injury was measured using single-molecule array (SIMOA) technology, which can track the minute blood levels of neurodegeneration markers in picograms (one trillionth of a gram) per millilitre of blood (pg/ml), where older technologies could not.
Three of the study markers—ubiquitin carboxy-terminal hydrolase L1 (UCHL1), total tau, and phosphorylated-tau-181 (ptau181)—are known measures of the death or disabling of neurons, the cells that enable nerve pathways to carry messages.
Levels of neurofilament light chain increase with damage to axons, which are extensions of neurons. Glial fibrillary acidic protein (GFAP) is a measure of damage to glial cells, which support neurons. Amyloid beta 40 and 42 are proteins known to build up in people who have Alzheimer’s disease.
Past study results argue that total tau and ptau181 are also specific measures of Alzheimer’s disease, but their role in the disease remains a matter of debate.
Blood markers in the COVID-19 patient group were measured in blood serum (the liquid part of blood that has been made to clot), while those in the Alzheimer’s disease study were measured in plasma (the liquid blood fraction that remains when clotting is prevented).
For technical reasons, the difference meant that neurofilament light chain, GFAP, and UCHL1 levels could be compared between the COVID-19 group and patients in the Alzheimer’s disease study, but total tau, ptau181, amyloid beta 40, and amyloid beta 42 could only be compared within the COVID-19 patient group (neuro symptoms or not; death or discharge).
Further, the main measure of neurological damage in patients with COVID-19 was toxic metabolic encephalopathy, or TME, with symptoms that range from confusion to coma, and caused during severe infections by toxins generated as the immune system overreacts (sepsis), kidneys fail (uremia), and oxygen delivery is compromised (hypoxia).
Specifically, the average percentage increase in levels of the seven markers for hospitalised patients with TME compared to those without neurological symptoms was 60.5 per cent.
For the same markers within the COVID-19 group, average percentage increase when comparing those successfully discharged home from the hospital to those who died in the hospital was 124 per cent.
A secondary set of findings came from comparing neurofilament light chain, GFAP, and UCHL1 levels in the serum of people with COVID-19 against levels of the same markers in the plasma of non–COVID-19 Alzheimer’s disease patients.
Neurofilament light chain was, over the short-term, 179 per cent higher (73.2 versus 26.2 pg/ml) in patients with COVID-19 than those with Alzheimer’s disease. GFAP was 65 per cent higher (443.5 versus 275.1 pg/ml) in patients with COVID-19 patients than those with Alzheimer’s disease, while UCHL1 was 13 per cent higher (43 versus 38.1 pg/ml).
Dementia protein changes discovered in terminal childhood disease
‘These findings challenge the view that changes to the brain that underlie these forms of dementia are merely age-associated’
A world-first study has revealed that protein changes seen in dementia are also present in a terminal childhood disease.
The pioneering research found that the brains of infants who passed away with an extremely rare genetic condition, Krabbe disease, have similar changes to those seen in two age-related brain diseases, dementia with Lewy bodies and Parkinson’s disease.
The study, from Newcastle University and funded by Alzheimer’s Research UK, shows alpha-synuclein has the potential to spread throughout the brain.
Dementia with Lewy bodies (DLB) and Parkinson’s disease are both neurodegenerative diseases associated with older age.
In both diseases, a protein called alpha-synuclein builds up in the brain into structures called Lewy bodies, causing damage to nerve cells.
In contrast, Krabbe disease is a rare inherited childhood disorder – with only one case in every 100,000 births.
The disease destroys the protective coating surrounding nerve cells in the brain and throughout the nervous system. It is caused by a mutation in the GALC gene, with symptoms showing within the first few months of life.
Sadly, the disease typically leads to death in those affected by the age of two.
The team at Newcastle studied postmortem brain tissue of four infants who died of Krabbe disease. While they did not identify Lewy bodies in the brain tissue of those infants, they did identify changes to the protein alpha-synuclein that are normally associated with DLB and Parkinson’s disease.
In particular, the researchers noted that the alpha-synuclein protein in Krabbe disease had the ability to stick together and spread through the brain, a key change associated with dementia with Lewy bodies and Parkinson’s disease that is thought to drive the progression of both diseases.
Dr Daniel Erskine, Alzheimer’s Research UK Research Fellow at Newcastle University, who is an expert in Lewy bodies and led the study, said: “Looking at the brains of four infants with Krabbe disease compared to four infants without Krabbe disease, we performed a series of experiments to establish the behaviour of the alpha-synuclein protein.
“We found that alpha-synuclein in Krabbe disease shared key qualities that are normally only associated with age-associated neurodegenerative disease like dementia with Lewy bodies.
“This phenomenon has only been reported previously in the brains of older people with neurodegenerative diseases, so seeing it in the brains of these infants is highly remarkable.
“These findings challenge the view that changes to the brain that underlie these forms of dementia are merely age-associated, but are instead the result of dysfunction of specific biological pathways.
“This is an important message, as while we cannot stop ageing, we can potentially fix something that is not working properly.
“These research findings have only been made possible from the support of the families donating their loved ones’ brains for research and I would like to pay tribute to them for their generosity in helping others in a time of unparalleled grief.
“Thankfully, Krabbe disease is a relatively rare condition, however it is devastating for the families affected. Dementia with Lewy bodies affects roughly 100,000 people in the UK.
“Building on these findings to help develop new drugs that target this common biological pathway could change the lives of people with both conditions.”
Dr Rosa Sancho, head of research at the charity Alzheimer’s Research UK, said: “In dementia with Lewy bodies, the protein alpha-synuclein forms clumps called Lewy bodies inside brain nerve cells.
“While Lewy bodies were not found in the brains of infants in this study, researchers have shown for the first time that alpha-synuclein had a similar capacity to spread through the brain in Krabbe disease.
“At Alzheimer’s Research UK we often hear dementia dismissed as a normal part of ageing.
“We know dementia is caused by brain diseases, and this research offers more evidence that the disease that cause dementia are caused by biological processes going awry rather than as a by-product of old age.
“Learning about the diseases that cause dementia from other neurodegenerative conditions and sharing this knowledge with other fields is important.
“We’re incredibly grateful to the families of those whose brains were studied in this research, and who made this important discovery possible.”
Insight4 weeks ago
Christmas is the best time of year – isn’t it?
Stroke4 weeks ago
‘Seek stroke treatment without delay’
Brain injury2 weeks ago
Work of Calvert Reconnections highlighted by BBC
Dementia4 weeks ago
Lonely this Christmas – the reality of living with dementia
Neuropsychology3 weeks ago
Fewer psychiatric beds linked to rise in prison numbers
News4 weeks ago
The Walton Centre appoints new non-exec directors
Brain injury1 week ago
Neumind: the app set to revolutionise ABI support
Insight2 weeks ago
It’s more than just colouring in!