Connect with us

Research

Machine learning could predict stages of Alzheimer’s

Published

on

Machine learning could predict stages of Alzheimer’s

Researchers in the US are exploring the use of machine learning to predict the advancement of Alzheimer’s disease and give a more accurate timeline of progression.

A new research collaboration has used machine learning to pinpoint the most accurate means, and timelines, for anticipating the advancement of Alzheimer’s disease in people who are either cognitively normal or experiencing mild cognitive impairment.

The modelling showed that predicting the future decline into dementia for individuals with mild cognitive impairment is easier and more accurate than it is for cognitively normal, or asymptomatic, individuals. 

At the same time, the researchers found that the predictions for cognitively normal subjects are less accurate for longer time horizons, but for individuals with mild cognitive impairment, the opposite is true.

The modelling also demonstrated that magnetic resonance imaging (MRI) is a useful prognostic tool for people in both stages, whereas tools that track molecular biomarkers, such as positron emission tomography (PET) scans, are more useful for people experiencing mild cognitive impairment.

The team’s paper, Machine Learning Based Multi-Modal Prediction of Future Decline Toward Alzheimer’s Disease: An Empirical Study, was published earlier this month. Lead author is Batuhan Karaman, a doctoral student in the field of electrical and computer engineering.

Alzheimer’s disease can take years, sometimes decades, to progress before a person exhibits symptoms. Once diagnosed, some individuals decline rapidly but others can live with mild symptoms for years, which makes forecasting the rate of the disease’s advancement a challenge.

“When we can confidently say someone has dementia, it is too late. A lot of damage has already happened to the brain, and it’s irreversible damage,” said senior author Mert Sabuncu, associate professor of electrical and computer engineering in the College of Engineering and of electrical engineering in radiology at Weill Cornell Medicine.

“We really need to be able to catch Alzheimer’s disease early on and be able to tell who’s going to progress fast and who’s going to progress slower, so that we can stratify the different risk groups and be able to deploy whatever treatment options we have.”

Clinicians often focus on a single ‘time horizon’ – usually three or five years – to predict Alzheimer’s progression in a patient. 

The timeframe can seem arbitrary, according to Sabuncu, whose lab specialises in analysis of biomedical data – particularly imaging data, with an emphasis on neuroscience and neurology.

Sabuncu and Karaman partnered with longtime collaborator and co-author Elizabeth Mormino of Stanford University to use neural-network machine learning that could analyse five years’ worth of data about individuals who were either cognitively normal or had mild cognitive impairment. 

The data, captured in a study by the Alzheimer’s Disease Neuroimaging Initiative, encompassed everything from an individual’s genetic history to PET and MRI scans.

The researchers discovered several notable patterns. For example, predicting a person will move from being asymptomatic to exhibiting mild symptoms is much easier for a time horizon of one year, compared to five years. 

However, predicting if someone will decline from mild cognitive impairment into Alzheimer’s dementia is most accurate on a longer timeline, with the “sweet spot” being about four years.

“This could tell us something about the underlying disease mechanism, and how temporally it is evolving, but that’s something we haven’t probed yet,” Sabuncu said.

Regarding the effectiveness of different types of data, the modelling showed that MRI scans are most informative for asymptomatic cases and are particularly helpful for predicting if someone’s going to develop symptoms over the next three years, but less helpful for forecasting for people with mild cognitive impairment. 

Once a patient has developed mild cognitive impairment, PET scans, which measure certain molecular markers such as the proteins amyloid and tau, appear to be more effective.

One advantage of the machine learning approach is that neural networks are flexible enough that they can function despite missing data, such as patients who may have skipped an MRI or PET scan.

In future work, Sabuncu plans to modify the modelling further so that it can process complete imaging or genomic data, rather than just summary measurements, to harvest more information that will boost predictive accuracy.

The research was supported by the National Institutes of Health National Library of Medicine and National Institute on Aging, and the National Science Foundation.

 

Research

Air pollution linked to increased hospital admission for heart and lung diseases

Published

on

Exposure to fine particulate matter (PM2.5) air pollution is linked to an increased risk of hospital admission for major heart and lung diseases, find two large US studies, published by The BMJ.

Together, the results suggest that no safe threshold exists for heart and lung health.

According to the Global Burden of Disease study, exposure to PM2.5 accounts for an estimated 7.6% of total global mortality and 4.2% of global disability adjusted life years (a measure of years lived in good health).

In light of this extensive evidence, the World Health Organization (WHO) updated the air quality guidelines in 2021, recommending that an annual average PM2.5 levels should not exceed 5 μg/m3 and 24 hour average PM2.5 levels should not exceed 15 μg/m3 on more than 3-4 days each year.

In the first study, researchers linked average daily PM2.5 levels to residential zip codes for nearly 60 million US adults (84 per cent white, 55 per cent women) aged 65 and over from 2000 to 2016. They then used Medicare insurance data to track hospital admissions over an average of eight years.

After accounting for a range of economic, health and social factors, average PM2.5 exposure over three years was associated with increased risks of first hospital admissions for seven major types of cardiovascular disease – ischemic heart disease, cerebrovascular disease, heart failure, cardiomyopathy, arrhythmia, valvular heart disease, and thoracic and abdominal aortic aneurysms.

Compared with exposures of 5 μg/m3 or less (the WHO air quality guideline for annual PM2.5), exposures between 9 and 10 μg/m3, which encompassed the US national average of 9.7 μg/m3 during the study period, were associated with a 29% increased risk of hospital admission for cardiovascular disease.

On an absolute scale, the risk of hospital admission for cardiovascular disease increased from 2.59% with exposures of 5 μg/m3 or less to 3.35% at exposures between 9 and 10 μg/m3.

“This means that if we were able to manage to reduce annual PM2.5 below 5 µg/m3, we could avoid 23% in hospital admissions for cardiovascular disease,” say the researchers.*

These cardiovascular effects persisted for at least three years after exposure to PM2.5, and susceptibility varied by age, education, access to healthcare services, and area deprivation level.

The researchers say their findings suggest that no safe threshold exists for the chronic effect of PM2.5 on overall cardiovascular health, and that substantial benefits could be attained through adherence to the WHO air quality guideline.

“On February 7, 2024, the US Environmental Protection Agency (EPA) updated the national air quality standard for annual PM2.5 level, setting a stricter limit at no more than 9 µg/m3. This is the first update since 2012. However, it is still considerably higher than the 5 µg/m3 set by WHO. Obviously, the newly published national standard was not sufficient for the protection of public health,” they add.*

In the second study, researchers used county-level daily PM2.5 concentrations and medical claims data to track hospital admissions and emergency department visits for natural causes, cardiovascular disease, and respiratory disease for 50 million US adults aged 18 and over from 2010 to 2016.

During the study period, more than 10 million hospital admissions and 24 million emergency department visits were recorded.

They found that short term exposure to PM2.5, even at concentrations below the new WHO air quality guideline limit, was statistically significantly associated with higher rates of hospital admissions for natural causes, cardiovascular disease and respiratory disease, as well as emergency department visits for respiratory disease.

For example, on days when daily PM2.5 levels were below the new WHO air quality guideline limit of 15 μg/m3, an increase of 10 μg/m3 in PM2.5 was associated with 1.87 extra hospital admissions per million adults aged 18 and over per day.

The researchers say their findings constitute an important contribution to the debate about the revision of air quality limits, guidelines, and standards.

Both research teams acknowledge several limitations such as possible misclassification of exposure and point out that other unmeasured factors may have affected their results. What’s more, the findings may not apply to individuals without medical insurance, children and adolescents, and those living outside the US.

However, taken together, these new results provide valuable reference for future national air pollution standards.

Continue Reading

Research

Home health care linked to increased hospice use at end-of-life – study

Published

on

Patients who had previously received home health care had a higher likelihood of accessing hospice care at the end of their life, according to a new study.

Researchers, whose findings are published in the Journal of Palliative Medicine, examined the home health care and hospice care experiences of more than two million people.

Using Medicare data, researchers found when individuals received home health care before the last year of their life, they had higher odds of using hospice care than those who had never received home health care.

Researchers said this association underscores the potential benefits of receiving end-of-life care in the comfort of one’s home.

As the aged population increases, the findings also show the need for more resources in the health care sector and staff training in end-of-life care.

Home health care services including skilled nursing, therapy, social work and aide services are used to maintain functioning or slow decline in health. Hospice care provides similar services but is intended for those with life expectancies of six months or less and is focused on pain relief, minimising hospital visits and providing comfort and support. Both services provide patients the opportunity to receive more personalised care in their home.

Researchers say home-based care also encourages greater involvement of family caregivers in the caregiving process.

Olga Jarrín, senior author of the study, the Hunterdon Professor of Nursing Research at the Rutgers School of Nursing and director of the Community Health and Aging Outcomes Laboratory within the Rutgers Institute for Health, Health Care Policy and Aging Research, commented: “In addition to benefits for the patient, hospice care also provides resources and support to help family caregivers cope with the physical, emotional and practical challenges of caring for a loved one at the end of life.”

Hyosin (Dawn) Kim, research assistant professor at Oregon State University and first author of the study, added: “By providing personalised care, reducing hospitalisations, fostering family involvement and support, and improving symptom management, home-based care can enhance the quality of end-of-life experiences for patients with terminal illnesses and their families.”

 

Continue Reading

Research

Sleep programme shows promise in those with memory problems – study

Published

on

A new study has shown promising results in improving sleep and quality of life in individuals living with memory problems.

A group of researchers from Penn Nursing, Penn Medicine, Rutgers School of Nursing, and Drexel University’s College of Nursing and Health Professions, have delved into the efficacy of a non-pharmacological approach in a trial known as the Healthy Patterns Sleep Program.

The study involved 209 pairings of community-residing individuals with memory problems and their care partners. Participants were assigned to either the Healthy Patterns Sleep Program, which consisted of one-hour home activity sessions administered over four weeks, or a control group that received sleep hygiene training, plus education on home safety and health promotion.

The Healthy Patterns Sleep Program trained care partners in timed daily activities such as reminiscence in the morning, exercise in the afternoon and sensory activities in the evening that can decrease daytime sleepiness and improve nighttime sleep quality.

Nancy Hodgson, PhD, RN, FAAN, the Claire M. Fagin Leadership Professor in Nursing and Chair of Department of Biobehavioral Health Sciences, who led the study, said: “The results from this study provide fundamental new knowledge regarding the effects of timing activity participation and can lead to structured, replicable treatment protocols to address sleep disturbances. Overall, the Healthy Patterns program resulted in improved QOL compared to an attention-control group.”

The findings also indicate that, compared to a control group, the four-week Healthy Patterns program improved sleep quality among persons living with memory issues who had depressive symptoms or poor sleep quality.  The study indicates the Healthy Patterns Intervention might need a longer dose to induce improvements in other sleep-wake activity metrics.

The study’s significance lies in its confirmation of the effectiveness of behavioural interventions in not only improving quality of life and addressing sleep quality issues in this population, but also potentially reducing care partner burden and overall care costs for persons living at home with memory problems.

Continue Reading

Trending