Connect with us


How protective genetic variant could fight Alzheimer’s



Five years ago, the global team of researchers identified a patient who did not develop cognitive impairment until her late 70s, despite being part of a family at extremely high genetic risk for developing early-onset Alzheimer’s disease.

Co-first author Yakeel T. Quiroz, of Massachusetts General Hospital (MGH), said: “As a clinician, I am highly encouraged by our findings, as they suggest the potential for delaying cognitive decline and dementia in older individuals. Now we must leverage this new knowledge to develop effective treatments for dementia prevention.

“As a neuroscientist, I’m thrilled by our findings because they underscore the complex relationship between APOE and a deterministic mutation for Alzheimer’s disease, potentially paving the way for innovative treatment approaches for Alzheimer’s disease, including targeting APOE-related pathways.”

Quiroz and her team at MGH and co-senior study author Joseph Arboleda-Velasquez, MD, PhD, of Mass Eye and Ear, have been working with their colleagues in Colombia as part of the MGH Colombia-Boston (COLBOS) biomarker study to examine family members of the world’s largest-known kindred with a genetic variant called the “Paisa” mutation (Presenilin-1 E280A).

The Paisa mutation is an autosomal dominant variant, meaning that inheriting just one copy of the mutated gene from a parent is enough to cause a genetic condition.

The family consists of about 6,000 blood relatives, and about 1,200 carry the variant. Carriers of this Paisa variant are destined to develop Alzheimer’s disease; most develop mild cognitive impairment  in their 40s, dementia in their 50s, and die from complications of dementia in their 60s.

Francisco Lopera, MD, director of the Grupo de Neurociencias de Antioquia in Medellín, Colombia, and co-senior author of the NEJM paper, is the neurologist who discovered this family and has been following them for the last 40 years.

In addition to the 2019 case report about a family member with two copies of the Christchurch variant, molecular studies have added further biological evidence that the variant could be playing a protective role.

In 2023, the research team identified another “resiliency gene variant” called Reelin-COLBOS that appeared to delay the onset of symptoms in other family members. The new study in NEJM reports on a larger group of individuals from this family who carry a copy of the Christchurch variant.

Co-senior author Joseph F. Arboleda-Velasquez said: “Our original study told us that protection was possible, and that was an important insight. But if a person needs two copies of a rare genetic variant, it just comes down to luck. Our new study is significant because it increases our confidence that this target is not only protective, but druggable. We think that therapeutics inspired by protected humans are much more likely to work and to be safer.”

The research team assessed 1,077 descendants of the Colombian family. They identified 27 family members who carried both Paisa mutation and one copy of the Christchurch variant.

On average, these family members began showing signs of cognitive impairment at age 52, compared to a matched group of family members who did not have the variant, who began showing signs at age 47. The family members also showed signs of dementia four years later than those who did not carry the variant.

Two of these individuals had functional brain imaging performed. Scans showed lower levels of tau and preserved metabolic activity in areas typically involved in Alzheimer’s disease, even in the presence of amyloid plaques — proteins considered a hallmark of Alzheimer’s disease.

The team also analysed autopsy samples from four deceased individuals that showed less pathology in blood vessels, a characteristic that appears important for the protective effects of APOE3 Christchurch.

The authors note that their study was limited to a relatively small number of people carrying both the Paisa and Christchurch variants, and to a single, extended family.

They write that further studies involving larger and more ethnically diverse samples of Alzheimer’s disease may shed further light on the protective effect of the Christchurch variant and help determine if findings from the family in Colombia could translate into discoveries relevant for treating sporadic forms of Alzheimer’s disease.

“As a next step, we are currently focused on improving our understanding of the brain resilience among the remaining family members who carry one copy of the Christchurch variant. This involves conducting structural and functional MRI scans and cognitive evaluations, as well as analyzing blood samples to assess their protein and biomarker profiles,” said Quiroz.

“The unwavering commitment to research shown by our Colombian patients with autosomal dominant Alzheimer’s and their families has been indispensable in making this study possible and allowing us to continue to work toward interventions for this devastating disease.”


The role of AI in early detection of dementia



By Mike Battista, Director, Science and Research, Creyos

Dementia is a debilitating condition affecting millions all over the world, characterised by a decline in cognitive functions such as memory, reasoning, and communication.

Early detection is advantageous in planning for and managing diseases that cause dementia, as early detection can identify mild issues with cognition while pathology in the brain is still minimal, and potential interventions have the greatest chance of slowing decline.

Early modification of lifestyle and medical risk have been estimated to prevent 40 per cent of dementia cases.

Artificial Intelligence (AI) is emerging as a ubiquitous tool in healthcare, offering new hope for early dementia detection through advanced data analysis and pattern recognition.

Dementia consists of a range of symptoms that can include memory loss, confusion, atypical behaviour, and difficulty completing everyday activities.

Mike Battista

Diagnosing dementia early is challenging with traditional methods, as so many of the most common diagnostic tools rely on significant cognitive decline.

However, early detection can lead to better management of dementia slowing the progression of symptoms and improving the overall quality of life for patients.

With these benefits in mind, innovative uses of AI are needed to create more sensitive tools that can detect the most subtle signs of dementia earlier than ever before.

AI in Identifying Dementia Markers

In relation to dementia, AI can search extensive datasets to find markers indicative of the condition.

Then, by analysing massive amounts of data, AI is advancing healthcare capabilities by identifying patterns indicative of the condition that clinicians might miss.

Understanding these data markers is imperative for accurate detection.

For instance, AI can analyse cognitive performance, language use, and even facial expressions to differentiate between healthy individuals and those showing early signs of dementia.

A strong understanding of these markers comes from reviewing scientific literature and adopting a data-driven approach.

For researchers, AI systems can be trained on large datasets to help discover the subtle differences in performance and behaviour associated with early dementia.

Clinicians can make use of the models trained by researchers to accurately identify patients with early signs of dementia.

Data from clinicians can also be used to further train AI models, improving the accuracy of early detection over time.

The Importance of High-Quality Data

High-quality, clean data are essential for AI to function effectively in detecting dementia.

These data must be accurate and comprehensive, containing the “truth” about individuals’ health status.

One of the biggest challenges in healthcare AI is obtaining this level of data, as it involves building a repository of information that includes both individuals with dementia and healthy individuals.

To overcome this challenge, there are various initiatives to gather diverse data.

These include gathering diagnostic information from patients and creating data partnerships between healthcare practices and research organisations.

These data partnerships are not only interesting for research purposes but also help gather high-quality real-world data from patient populations that can be used to train AI systems for practical clinical application.

Longitudinal datasets are also key to achieving early diagnosis.

They are required to answer research questions about which early signs progress into dementia, which lifestyle factors slow progression over time, and more.

However, gathering these datasets is a complex process.

It requires collecting data from a large group of healthy individuals at a specific time and tracking who develops conditions like mild cognitive impairment or dementia over time.

This involves at least two time points with good follow-up, minimal dropout, and accurate diagnoses, making it a complicated dataset to get right.

Additionally, there is a great deal of selection bias in the patients who are repeatedly tested.

That is because these patients are likely more proactive about addressing their health, and/or more likely to already have issues and seek medical attention because of concerns about their cognition.

What is needed are large scale, randomised controlled studies where people are recruited ahead of time and followed throughout their health journey with state-of-the-art tools to assist in the process.

For example, the Maintain Your Brain trial followed thousands of older participants over three years to examine the effectiveness of lifestyle interventions meant to slow cognitive decline.

They used online cognitive testing from Creyos to ensure that testing was easy for participants, but provided high-quality data for the researchers.

normative dataset from healthy individuals allows researchers to compare individuals with a known condition to the general population, ensuring AI models can accurately distinguish the two groups.

Building and maintaining these datasets is a complex process but is critical for the success of AI applications.

An example of successful implementation of machine learning is the Creyos dementia screener, which used data from thousands of neurology clinic patients and a health normative database to train a model that accurately detects individuals with subtle cognitive impairments.

Current Success and Future Potential

Several success stories illustrate AI’s potential in early dementia detection.

For instance, pilot programmes using AI have shown promising results in identifying early cognitive decline through routine health assessments and patient interactions.

Examples of how AI recorded observations include:

  • How the patient moved the computer mouse while performing assessments
  • The time taken by the patient to read the tutorial
  • Passive data gathering such as device motion data (noting a person’s gait) handwriting, virtual reality integration (to test for driving ability), facial expression and speech pattern analysis.

These pilot programmes demonstrate novel ways to exploit AI’s capability to provide real-time monitoring and personalised treatment plans, tailored to individual patients’ needs.

It’s clear that the future potential of AI in dementia care is immense.

As technology advances, AI could enable even more precise and individualised healthcare, offering the earliest interventions and continuous monitoring that could further slow the progression of dementia.

This proactive approach could revolutionise how we manage and treat dementia, improving patient outcomes.

Ethical Considerations and Challenges

While AI offers great potential, it also raises ethical considerations and challenges.

Privacy concerns regarding patient data must be addressed, ensuring that data is securely handled, and patients’ confidentiality is maintained.

Additionally, the bias that can get baked into some AI systems must be scrutinised to maintain accuracy, and avoid false positives and negatives, which could lead to misdiagnosis and inappropriate treatments.

Ethical use of AI in healthcare involves balancing technological advancements with patient rights.

Furthermore, AI cannot replace human decision-making. Machine learning models and AI software can enhance human judgment, but a physicians must always make the final call when it comes to diagnosing dementia

AI holds tremendous promise in the early detection of dementia, offering tools that can analyse complex data and identify subtle signs of disruptions to cognition.

To fully realise this potential, continued research and investment in AI technologies and high-quality data collection are essential.

About the Author – Mike Battista, Director, Science and Research, Creyos 

Mike Battista is the Director of Science and Research at Creyos.

His interests and science communication focus on brain health, cognition, and neuropsychological testing.

He received his PhD in personality and measurement psychology at Western University in 2010 and has been exploring the intersection of science and technology ever since.

For more information about Creyos Health visit 

Continue Reading


Opinion: How robots could help to ease the social care crisis



Robear is an animatronic bear that lifts elders with mobility problems, Paro is a fuzzy robotic seal intended to provide a futuristic form of animal therapy, and Pepper is a humanoid with remote-monitoring capabilities and learning tools.

Pet robots have revolutionised and enhanced the standard of care, improving the wellbeing of care home residents. Here Stephen Hayes, managing director at automation tech firm Beckhoff UK, explores the benefits of integrating robots in social care homes.

There is a wide variety of care robots on the market. Some are aimed at physical care, including machines that can assist with mobility and exercise, feed their owner and help them with hygiene tasks. These could greatly benefit caregivers, freeing their time and preventing them from suffering from long-term health conditions or disability due to the physical effort associated with giving care.

Others play the role of a companion, engaging older people emotionally to reduce and even prevent cognitive decline, providing companionship for lonely older people, and making those with cognitive conditions easier for care staff to manage.

Research by the University of Plymouth, conducted in care homes using these pets found decreased neuropsychiatric symptoms such as delusions, depression, anxiety, apathy and occupational disruptiveness because they provided a sense of responsibility and purpose.

Social care vacancies are higher than before the COVID-19 pandemic, and data shows one in ten social care posts are unfilled in a staffing crisis that could have harmful results for residents. In England, 152,000 social care posts remain empty, according to a report released by Skills for Care.

Steve Barclay, Secretary of State for Health and Social Care commented for The Telegraph earlier this year, stating that robots and AI are key to better supporting patients and reducing demand on social care staff. He said that there was a need to adopt an innovative approach to health and attempt to cut NHS waiting times while improving care for the elderly.

However, these robots present limitations, such as superficiality and lack of personalisation. Also, the content of their conversations can be very limited, making them less entertaining with the pass of time.

This was the case of the humanoid Pepper, for which production ceased in 2021 due to a weak demand as care homes did not see the long-term benefit of his interactions. Nevertheless, robots like Paro, which move and respond to touch, have had a positive impact on the wellbeing of care home residents. However, with a cost of £5,000, care homes are looking for more affordable options.

Japan is a pioneer in developing this kind of technology. The nation is facing a ‘greying’ crisis due to the aging of its population, so the country has invested heavily in developing caretech able to serve and provide emotional support.

In the UK, there are currently almost 12 million people who are aged 65 or over, and the number of people coping with illnesses such as arthritis or dementia is expected to increase. In fact, a recent machine learning study by the Journal of Medical Systems suggests that 135 million people might be affected by dementia by 2050.

To allow for continuous innovation in this field, the UK Government announced its commitment to invest at least 2.4 per cent of GDP in R&D by 2027. The programme supports the UK Government’s Ageing Society Grand Challenge and Future of Mobility Grand Challenge, which will ensure we meet the needs of an ageing society.

With this in mind, there are certain technologies that we are likely to see more of in care homes over the coming years, including robots that can connect to each other and other devices.

This includes devices such as oximeters, thermometers, or even thermal cameras, enabling the elderly to have consultations any time of the day, from home, and send out emergency notices to staff or hospitals. However, for care robots to be a success, state-of-the-art control technology is required.

Beckhoff’s Ethernet-based fieldbus system, EtherCAT, has extension modules that are compatible with third party hardware for integration. This platform process data and transports it directly, has a flexible topology and simple line or tree structure that requires no expensive infrastructure components and includes the environments for programming, diagnostics and configuration.

This global standard for real-time Ethernet communication provides workers with real-time information about elders like location, health condition, or learning progress. This data also allows carers focus their time on other urgent tasks, optimise resources and personalise treatment.

The advancement and implementation of robot pets could improve awareness of preventative care, reduce anxiety on disease and enhance stakeholder relationship.

Further research on caretech would tackle functional problems, making these devices an essential asset for any caregiver. By investing in the right control technology now, social care homes will be better prepared to take care of their residents.

Beckhoff provides PC-based control and EtherCAT to connect caretech systems. See more here.

Continue Reading


How AI speech analysis could predict dementia onset



AI could be used to predict whether a person will develop Alzheimer’s-associated dementia by simply analysing their speech, scientists believe.

Researchers at Boston University are exploring how analysis of speech patterns via a machine learning model could detect with a high degree of accuracy whether someone with mild cognitive impairment will develop Alzheimer’s-associated dementia within six years

They say their model can predict, with an accuracy rate of 78.5 percent, whether someone with mild cognitive impairment is likely to remain stable over the next six years—or fall into the dementia associated with Alzheimer’s disease.

While allowing clinicians to make earlier diagnoses, the researchers say their work could also help make cognitive impairment screening more accessible by automating parts of the process;  with no expensive lab tests, imaging exams, or office visits required.

Ioannis (Yannis) Paschalidis, director of the Boston University Rafik B. Hariri Institute for Computing and Computational Science & Engineering, says: “We wanted to predict what would happen in the next six years—and we found we can reasonably make that prediction with relatively good confidence and accuracy.

“We hope, as everyone does, that there will be more and more Alzheimer’s treatments made available.

“If you can predict what will happen, you have more of an opportunity and time window to intervene with drugs, and at least try to maintain the stability of the condition and prevent the transition to more severe forms of dementia.”

The project involves a multidisciplinary team of engineers, neurobiologists, and computer and data scientists.

“We hope, as everyone does, that there will be more and more Alzheimer’s treatments made available,” says Paschalidis.

“If you can predict what will happen, you have more of an opportunity and time window to intervene with drugs, and at least try to maintain the stability of the condition and prevent the transition to more severe forms of dementia.”

To train and build their new model, the researchers turned to data from one of the oldest and longest-running studies in the US —the BU-led Framingham Heart Study.

Although the Framingham study is focused on cardiovascular health, participants showing signs of cognitive decline undergo regular neuropsychological tests and interviews, producing a wealth of longitudinal information on their cognitive well-being.

Paschalidis and his colleagues were given audio recordings of 166 initial interviews with people, between ages 63 and 97, diagnosed with mild cognitive impairment—76 who would remain stable for the next six years and 90 whose cognitive function would progressively decline.

They then used a combination of speech recognition tools—similar to the programs powering your smart speaker—and machine learning to train a model to spot connections between speech, demographics, diagnosis, and disease progression.

After training it on a subset of the study population, they tested its predictive prowess on the rest of the participants.

“We combine the information we extract from the audio recordings with some very basic demographics—age, gender, and so on—and we get the final score,” says Paschalidis. “You can think of the score as the likelihood, the probability, that someone will remain stable or transition to dementia. It had significant predictive ability.”

Rather than using acoustic features of speech, like enunciation or speed, the model is just pulling from the content of the interview—the words spoken, how they’re structured.

And Paschalidis says the information they put into the machine learning program is rough around the edges: the recordings, for example, are messy—low-quality and filled with background noise.

“It’s a very casual recording,” he says. “And still, with this dirty data, the model is able to make something out of it.”

That’s important, because the project was partly about testing AI’s ability to make the process of dementia diagnosis more efficient and automated, with little human involvement.

In the future, the researchers say, models like theirs could be used to bring care to patients who aren’t near medical centers or to provide routine monitoring through interaction with an at-home app, drastically increasing the number of people who get screened.

According to Alzheimer’s Disease International, the majority of people with dementia worldwide never receive a formal diagnosis, leaving them shut off from treatment and care.

Rhoda Au, a coauthor on the paper, says AI has the power to create “equal opportunity science and healthcare.”

The study builds on the same team’s previous work, where they found AI could accurately detect cognitive impairment using voice recordings.

In future research, Paschalidis would like to explore using data not just from formal clinician-patient interviews—with their scripted questions and predictable back-and-forth—but also from more natural, everyday conversations.

He’s already looking ahead to a project on if AI can help diagnose dementia via a smartphone app, as well as expanding the current study beyond speech analysis—the Framingham tests also include patient drawings and data on daily life patterns—to boost the model’s predictive accuracy.

“Digital is the new blood,” says Au. “You can collect it, analyse it for what is known today, store it, and reanalyse it for whatever new emerges tomorrow.”

Continue Reading