Non-Invasive Alzheimer’s Breakthroughs Transforming Detection

More than 7 million Americans are living with Alzheimer’s today.  This number is only going to increase in the coming years, projected to rise to nearly 13 million by 2050, as per the data provided by the Alzheimer’s Association.

Similarly, the health and long-term care costs for people living with dementia are forecasted to reach almost $1 trillion in 2050.

Globally, over 57 million people were living with dementia in 2021, with Alzheimer’s disease (AD) accounting for most (60-70%) of cases. These figures are expected to reach 139 million by 2050. 

When it comes to the lifetime risk for Alzheimer’s at age 45, it is particularly high for women, 1 in 5, compared to 1 in 10 for men. Also, two-thirds of Americans with Alzheimer’s are women.

Moreover, older Black Americans are about twice as likely to have the disease as older Whites, while older Hispanics are about one and a half times as likely.

While Alzheimer’s disease most often happens after age 65, with the risk increasing significantly with age, it can also occur in people younger than this in cases of rare early-onset Alzheimer’s. 

Researchers believe¹ about 110 of every 100,000 people aged 30 to 64 years have younger-onset dementia.

So what exactly is Alzheimer’s disease? Well, it is a progressive brain disorder that slowly destroys memory, thinking, reasoning, and learning skills. It can also change behavior and personality, as well as weaken or eliminate language and spatial understanding.

This neurodegenerative disease is the most common form of dementia, a general term for cognitive decline and memory loss.

Alzheimer’s disease is not a normal part of aging. It is a biological process that starts with the buildup of proteins in the brain. Two proteins in particular are responsible here:

1. Amyloid, which forms plaques outside of brain cells
2. Tau, which forms tangles inside brain cells

The buildup of these two proteins causes the death of brain cells and leads to brain shrinkage.

Early symptoms of this disease include forgetting recent conversations or events, and then, over time, leading to serious memory loss that affects a person’s ability to do even simple everyday tasks.

In advanced stages, the loss of brain function can even cause infection, dehydration, or poor nutrition, potentially resulting in death. Alzheimer’s actually kills more people than breast and prostate cancer combined. For years now, it has been among the ten leading² causes of death in people aged 65 and older.

There is, however, no cure for Alzheimer’s disease. What medicines do is, they help slow its progression and manage symptoms.

Now, to diagnose this disease, healthcare providers use the patient’s health, medical history, daily routine, and any changes in the patient’s mood and behavior. They also use a few tests, including blood tests, cognitive tests, CSF tests, brain MRI, CT scan, PET scan, and psychiatric and mental health evaluations.

Research in this space is expanding the range of these tests and tools for early detection, including using AI and advanced imaging to identify changes in the brain. 

Non-invasive breakthroughs for Alzheimer’s are also being explored for early detection and diagnosis, as well as addressing symptoms.

In regard to that, just earlier this year, a clinical trial found that repetitive focused ultrasound-mediated blood-brain barrier (BBB) opening on the frontal lobes is not only safe but can also reduce amyloid plaques.

It also improved common neuropsychiatric symptoms (anxiety, agitation, irritability, and delusions) related to the disease. According to Neal Kassell, MD, founder and chairman of the Focused Ultrasound Foundation, which funded the research led by Korea University Anam Hospital in collaboration with Yonsei University:

“Alzheimer’s research has remained relatively stagnant over the past few decades, but focused ultrasound offers hope in a field that has long sought innovative solutions and has the potential to disrupt the course of this devastating disease.“

Click here to learn how AI can help predict Alzheimer’s disease.

Resting Heart Rate Improves Dementia Risk Prediction Across Diverse Populations

While scientists have begun looking beyond the brain for early warning signs, there is one obvious connection that’s being missed: the heart-brain connection. An international team of researchers has actually found that resting heart rate can help with the detection of dementia risk with increased accuracy across most racial groups.

Heart rate is simply the number of times our heart beats every minute. A normal heart rate at rest, which is when we are calm and inactive, ranges between 60 and 100 beats per minute. Changes in heart rate can suggest a heart condition or other health problems.

The same heartbeat measurement that helps us monitor our heart health and fitness level, according to researchers, can also help predict the risk of dementia. 

According to Newman Sze, the Professor of Health Sciences at Brock University and the Canada Research Chair in Mechanisms of Health and Disease, abnormal heart rate is one of the most important risk factors for dementia after obesity and hypertension. 

For instance, they may signal underlying chronic stress and autonomic dysfunction, which potentially contribute to neurodegeneration and poor cerebral perfusion.

“If the resting heart rate is too low or too fast due to heart muscle failure, there’s not enough blood being pumped to the brain. The brain doesn’t receive enough oxygen and nutrients, which leads to brain degeneration.”

– Sze

The feature, however, is not captured in one of the most widely used prognostic tools, the CAIDE model.

In order to assess a patient’s susceptibility to developing dementia in the future, the Cardiovascular Risk Factors, Aging and Incidence of Dementia (CAIDE) international evaluation tool uses several physiological and social measurements.

The CAIDE model has been fundamental in clinical decision-making, patient counseling, and risk management.

While showcasing strong predictive capabilities, the current model does not capture the complete picture of a person’s health, especially across diverse racial groups in the U.S., said. The limited validity of the existing model may translate into unequal access to health care, disparities in quality of care, and variations in dementia-related risk factors, such as cardiovascular disease. 

Moreover, the prediction models, as the study noted, are often built using a very selective population, which doesn’t work for the diverse demographics. 

So, the eight-member research team investigated the impacts of including resting heart rate (RHR) in the CAIDE model to see if adding this would improve the model and increase equitable access to dementia prediction.

RHR, after all, is an accessible and non-invasive marker of heart health, which, unlike traditional cardiovascular risk factors, offers additional information related to autonomic nervous system function and cardiovascular stress responses.

To assess RHR’s effectiveness, the team used the data of 44,467 US participants aged 18 and older, including those aged 65 and above, collected by the National Alzheimer’s Coordinating Center (NACC) between 2005 and 2023. It also included information from cognitive tests, physical examinations, and interviews.

To develop the model, the team used a random forest algorithm in the NACC dataset.

The machine learning (ML) technique captured complex, non-linear relationships among variables, enhancing the team’s dementia risk prediction. 

Participants in the database were first divided into self-reported racial groups: White, Black African, Hispanic, Asian, and two indigenous populations: American Indian and Alaska Native.

The research team then ran each group through the current CAIDE model, consisting of age, sex, level of education, physical activity, body mass index (BMI), cholesterol level, and hypertension measurements, along with the apolipoprotein E (APOE) ε4 allele biomarker, the strongest genetic risk factor for AD.

The process was then repeated with the CAIDE-RHR model that included resting heart rate. Sze says:

“This adjustment significantly improved dementia risk prediction across most racial groups, offering a more inclusive and accessible way to identify at-risk individuals.” 

The good thing about this is that resting heart rate is easy to measure, which means more people can be screened and monitored. And that makes the CAIDE-RHR model more inclusive.

While attempts have been made in the past to improve the CAIDE model’s accuracy through resource-intensive lab analysis to detect dementia biomarkers in blood samples, this poses the risk of reducing access for multi-racial, underserved populations.

“In contrast, resting heart rate can be measured with a simple blood pressure cuff or by placing fingers on the wrist — methods that are quick, non-invasive and widely available, even in underserved community settings.”

– Shakiru Alaka, a PhD student and the lead author of the study.

According to the findings of the study, which was published³ in Alzheimer’s & Dementia: The Journal of the Alzheimer’s Association, the CAIDE-RHR model improved the accuracy of dementia risk prediction significantly for all racial groups in the study except the American Indigenous populations. But that could be due to the low number of participants, it noted. According to Sze:

“This finding highlights the important connection between heart health and brain health.”  

The low-cost, non-invasive CAIDE-RHR model, he continued, is “a step toward addressing systemic gaps in how we assess dementia risk across diverse populations” that could be integrated into routine care “to help identify those at risk earlier and more equitably.”

Video-Based Cognitive Tasks Spot Alzheimer’s Years Before Symptoms

While the previous study used ML and resting heart rate, a study from Rutgers-Newark researchers is using video games for early detection of Alzheimer’s and has found it to be as effective as blood tests.

The researchers developed the games to help detect the disease years before its symptoms first become noticeable.

These innovative dementia tests provide a non-invasive form of dementia screening, matching the results of widely available blood tests that reveal biomarkers for Alzheimer’s disease. The new non-invasive tests actually come with the added advantage of being painless and cost-effective because they do not require trained technicians to obtain blood samples.

Published in the journal Alzheimer’s Research & Therapy, the study⁴ was conducted by the Aging & Brain Health Alliance, which focuses on the role of genetics and lifestyle in delaying Alzheimer’s. 

With the latest discovery, researchers won’t have to wait for the disease to be conventionally diagnosed and can choose drug trial participants in the earliest stages of the disease in a noninvasive way, saving years in the clinical trial timetable.

“It’s pretty exciting for us because even before any problems with cognition become obvious, we have an early warning sign.”

– Lead author Miray Budak of the Center for Molecular & Behavioral Neuroscience

The alliance has actually been developing and testing its video-game screening tool for over two decades now. And finally, they have more evidence confirming that it is, in fact, effective at detecting reduced brain function years before patients or those close to them notice any symptoms.

The video-game test, which is called a generalization task, measures a person’s cognitive ability. That is done by demonstrating how well a person can work out a rule related to shapes and colors and then apply that to new examples. 

The team also developed a different assessment that uses MRI imaging to detect declines in brain flexibility.

To test the effectiveness of their tools, the team recently conducted a study involving 148 participants, all of whom were African American and cognitively unimpaired. They took several cognition tests, including the generalization task, before giving blood samples and undergoing a brain MRI.

Unlike the cognition tests currently in use, which require participants to recall a list of words or draw a clock face and often fail to detect Alzheimer’s symptoms until it’s too late, the Rutgers-Newark tools are simple, globally usable, and support early intervention, allowing people to be more protective of their brain health.

Click here to learn how smell-based VR therapy can help slow cognitive decline.

Personalized Digital Twin Models for Early Alzheimer’s Diagnosis

A new study, meanwhile, has created a digital twin model to help with the diagnosis of Alzheimer’s disease in its preclinical stages, like subjective cognitive decline (SCD), which can allow for timely management of the condition.

The current early diagnostic methods are unsuitable for such preclinical screenings due to their limited availability and diagnostic reliability. Also, they rely on invasive and scarcely available methods, which increases AD underdiagnosis in its preclinical forms.

To overcome these challenges, the study⁵ published in Alzheimer’s Research & Therapy presented the Digital Alzheimer’s Disease Diagnosis or DADD model, which provides digital biomarkers of Alzheimer’s disease by leveraging personalized brain modeling and EEG recordings. EEG signals were collected from healthy as well as SCD patients.

Electroencephalography (EEG) is a widely adopted tool that has been extensively used to investigate the effects of AD and cognitive decline in the electrical brain activity measured during cognitive tasks or in the resting state. 

While EEG comes with the benefits of reduced costs and wide availability, unlike more expensive diagnostic methods like MRI and PET scans, it does have its limitations, mainly concerning spatial resolution and volume conduction effects.

Also, there’s no diagnostic pipeline for AD based on EEG recordings that has reached clinical use.

Here, the study noted that computational models and digital twins offer a promising solution, but they are scarcely used in clinical settings.

So, the team created the DDD model, which used mechanisms associated with the diseases to create a personalized digital twin for each patient. The DADD model has shown high accuracy in forecasting cerebrospinal fluid (CSF) biomarkers of Alzheimer’s as well as conversion to clinical cognitive decline.

Digital biomarkers derived from the model were able to robustly distinguish between SCD and healthy participants, with a 7% improvement in classification accuracy compared to standard EEG biomarkers.

The model also successfully identified patients who were positive for CSF biomarkers of AD with 88% accuracy, much higher than the 58% accuracy of EEG biomarkers. The study noted:

“Predicting CSF biomarkers by combining digital twins with non-invasive recordings could revolutionize AD diagnosis in its early stages, paving the way for the clinical application of digital twins in AD diagnostics.”

Magnetic Sensor Finger-Tapping Tests Reveal Alzheimer’s in Its Earliest Stage

While cognitive impairment is a core and early symptom of Alzheimer’s disease, it can also change the way the body works.

These physical changes could be stiff muscles, fatigue, loss of balance or coordination, dragging feet, trouble standing or sitting up, and uncontrollable twitches.

Studies that looked into finger function in dementia patients have found deteriorations in fine motor control. Also, they have reported longer intervals between finger-tapping and fewer taps in AD and Mild Cognitive Impairment (MCI) patients compared to healthy elderly individuals.

To test that, Japan’s National Center for Geriatrics and Gerontology (NCGG) and Hitachi collaborated and reported a high correlation between specific finger-tapping movement and Alzheimer’ s-type dementia. For this, they used a waveform analysis technique that allowed a variety of tapping patterns to be extracted from the motor ability data with the help of magnetic sensors (UB1).

Patients with dementia displayed slower and less regular tapping, demonstrating that this simple, rhythmic movement can serve as an early indicator of cognitive decline. 

In their subsequent joint research, NCGG and Hitachi explored⁶ differences in finger movements during finger-tapping among healthy elderly individuals and AD and MCI patients. This time, they used UB-2, the improved magnetic sensing finger-tap device.

According to the study, patients with Alzheimer’s disease may experience delayed contact duration during finger-tapping, irregular rhythm, and time lag between the two hands.

The study results showed that the contact duration in AD and MCI patients was “significantly longer” than that in healthy elderly individuals. This delay in contact duration during finger-tapping, it is noted, might be a characteristic pattern observed from the MCI stage, as an earlier stage.

Swipe to scroll →

Investing in Alzheimer’s Treatment  

The global pharmaceutical company, Eli Lilly and Company , is a prominent name in the sector, actively developing Alzheimer’s drugs. 

Last year, its Alzheimer’s treatment for adults with early symptomatic AD, Kisunla™ (donanemab-azbt), received FDA approval. The treatment has shown promising results in slowing cognitive decline in early-stage patients

Eli Lilly and Company

Eli Lilly is a $605.2 billion market cap company, whose shares are currently trading at $648.38, down 17.17% YTD. The company’s stock hit its peak at $972.5 late in 2024. 

The stock recently experienced a crash to a 19-month low, which was due to an unexpected setback for its oral obesity drug, orforglipron. While patients who took the highest dosage of the pill for 72 weeks lost up to 11.5% more body weight than those who took a placebo, the results were weaker than Novo Nordisk‘s (NVO +0.39%) Wegovy.

Eli Lily’s second quarter results, however, painted a bullish picture. Its sales were up 38% to $15.56 bln while adjusted earnings per share were up 61% to $6.31.

Latest Eli Lilly (LLY) Stock News and Developments

The Future of Non-Invasive Alzheimer’s Detection

Every three seconds, someone develops dementia, which shows just how critical it is to find effective treatments for this debilitating disease. But while there may not be a cure yet, researchers are achieving powerful breakthroughs.

From pulse readings to finger taps, and gaming to digital brain twins, Alzheimer’s research is converging toward faster, cheaper, and more inclusive diagnostics. By detecting and managing the symptoms long before they surface, these everyday measures can enable people to preserve their cognitive health for longer and reduce the burden on caregivers!

Click here to learn how advances in AI and CRISPR technologies will help expedite our understanding of genetic diseases.

References:

1. He, Q., Wang, W., Zhang, Y., Xiong, Y., Tao, C., Ma, L., You, C., Ma, J., & Jiang, Y. Global burden of young-onset dementia, from 1990 to 2021: an age-period-cohort analysis from the Global Burden of Disease Study 2021. Translational Psychiatry, 15(1), 56, published 17 February 2025. https://doi.org/10.1038/s41398-025-03275-w
2. Alzheimer’s Association. 2022 Alzheimer’s disease facts and figures. Alzheimer’s & Dementia, 18(4), 700–789, published 14 March 2022. https://doi.org/10.1002/alz.12638
3. Alaka, S.A., Ngan, S.-F.C., Shookoni, M., MacPherson, R.E.K., Faught, B.E., Klentrou, P., Kalaria, R., Chen, C.P., & Sze, S.K. Enhancing the validity of CAIDE dementia risk scores with resting heart rate and machine learning: An analysis from the National Alzheimer’s Coordinating Center across all races/ethnicities. Alzheimer’s & Dementia, published 8 August 2025. https://doi.org/10.1002/alz.70442
4. Budak, M., Fausto, B.A., Osiecka, Z., et al. Elevated plasma p-tau231 is associated with reduced generalization and medial temporal lobe dynamic network flexibility among healthy older African Americans. Alzheimer’s Research & Therapy, 16, 253, published 22 November 2024. https://doi.org/10.1186/s13195-024-01619-0
5. Amato, L.G., Lassi, M., Vergani, A.A., et al. Digital twins and non-invasive recordings enable early diagnosis of Alzheimer’s disease. Alzheimer’s Research & Therapy, 17, 125, published 31 May 2025. https://doi.org/10.1186/s13195-025-01765-z
6. Sugioka, J., Suzumura, S., Kawahara, Y., Osawa, A., Maeda, N., Ito, M., Nagahama, T., Kuno, K., Shiramoto, K., Kizuka, S., Mizuguchi, T., Sano, Y., Kandori, A., & Kondo, I. Assessment of finger movement characteristics in dementia patients using a magnetic sensing finger-tap device. Japanese Journal of Comprehensive Rehabilitation Science, 11, 91–98, published 2020. https://doi.org/10.11336/jjcrs.11.91