How AI can power early detection of dementia

"Neurocognitive health interventions must be multipronged and personal, which necessitates the use of digital tools to adequately equip health systems to meet this growing need," one expert says.
By Bill Siwicki
09:52 AM

David Bates, CEO and cofounder of Linus Health

Photo: Linus Health

Biogen recently announced it is eliminating Adulhelm spending, which brings into question whether drugs are the only answer to Alzheimer's – especially when advanced, IT-enabled early detection technologies and interventions are available.

Cognitive screening methods supported by technology are growing in popularity as healthcare professionals realize the value of having faster, more affordable access to cognitive data that informs actionable next steps, says David Bates, CEO and cofounder of Linus Health, a brain health technology company.

Healthcare IT News sat down with Bates to discuss the potential of early detection via these sophisticated screening methods and explain how AI-powered testing, using widely available devices, shows decline earlier than a human eye can detect.

Q. What kinds of IT-enabled early detection technologies and interventions are available for treating Alzheimer's?

A. Healthcare providers have long relied on paper and pencil-based tests to detect early signs of cognitive impairment, dementia and Alzheimer's disease in their patients and serve as the basis of referrals to specialists.

Similar to so many other parts of healthcare, though, digital innovation and device accessibility have opened up a wealth of opportunities to improve how providers detect and address brain health issues in primary care.

As the global population ages and we contend with daunting projections about a commensurate rise in cognitive and brain disorders, but not a commensurate increase in the number of specialists to help render care, digital technology offers new promise for earlier detection and timely intervention in the primary care setting.

Digitizing and expanding longstanding paper tests has been a key step in making cognitive assessments more objective, practical and efficient. Among numerous other benefits, digital solutions can simplify administration with standardized, tech-enabled processes; streamline scoring and enhance interrater reliability through automation; and facilitate both historical tracking and care collaboration with patients' results accessible online.

More advanced digital solutions go even further, not only providing insights based on final test outputs, but also utilizing AI to assess a patient's whole process of taking a test for greater test sensitivity. This is particularly relevant because, in practice, assessment of the process – as much as or even more than the result – is what specialists like neurologists and neuropsychologists use to formulate a clinical impression about a patient's health.

What also sets more advanced solutions apart is the "so what?" portion. An assessment score is only the beginning. From there, providers – particularly primary care providers who first diagnose the majority of dementia cases – need actionable information to help them navigate what to do next for their patients, especially as the world continues to await breakthrough drugs.

While there may not be a surefire "treatment" yet, early detection affords patients the possibility of joining clinical trials and pursuing the best other options available today, which are evidence-based lifestyle and health adjustments.

There is a growing body of evidence supporting the latter. For example, in 2020, the Lancet Commission reported that modifying 12 risk factors, such as physical inactivity or smoking, may prevent or delay up to 40% of dementia cases. However, we are not fully capitalizing on these types of health and lifestyle interventions in brain health today, especially as they often come into play too late to maximize their impact.

Q. You suggest cognitive screening methods supported by technology are growing in popularity as healthcare professionals realize the value of having faster, more affordable access to cognitive data that informs actionable next steps. Please elaborate.

A. Primary care providers are essential to our healthcare system and deeply committed to the care of their patients – they're also stretched incredibly thin. While specialist shortages are often in the headlines, they are mounting in primary care, too, which the American Medical Association underscored in a new report just this past April.

To make their limited time with patients as effective as possible, any novel process or technology introduced into practice workflows has to improve on the status quo in terms of both clinical care and operational efficiency, while also having a reimbursement path for the additional time and workload. As these providers' communities rely on them even more heavily to counter the growing tide of cognitive decline cases, empowering them with new tools is essential.

Technology-based solutions can produce the dual benefit of improving care and operational efficiency in a number of ways. For example, digital solutions can make screening more practical for practices by simplifying training, data capture and scoring.

In addition to making it easier for staff to administer assessments, digital solutions also give providers quick, online access to results to discuss with their patients. From there, with solutions that include built-in guidance, providers also have helpful guidance at their fingertips as they determine next steps – an immense benefit simply not possible with paper-based tests that provide only static information on a page.

Lastly, CMS has now made traditional neurology CPT codes – for example, 99483 – available to primary care providers and encouraged their use. Realizing the potential of such codes all but necessitates the use of digital tools to satisfy the comprehensive requirement criteria.

The potential benefits of digital screening methods don't stop at what's possible in a provider's office or other clinical setting, though. Solutions accessible on widely available devices like tablets and smartphones also offer the ability to expand remote screening and monitoring of interventions, such as from a patient or caregiver's home.

In keeping with the broader growth in alternative care settings and tech-enabled appointment preparation, these capabilities can help providers further preserve and maximize in-clinic time for the more important aspects of results interpretation, consultation and care plan development.

Q. How can AI-powered testing, using widely available devices, show decline earlier than a human eye can detect?

A. AI and machine learning often are positioned as a means of replacing or reducing reliance on humans, but that's not how we view the role of AI when it comes to brain health. In brain health, AI is unlocking a whole world of possibility for earlier detection – and intervention – at a time when the urgency has never been greater to find different approaches that can change the global outlook on brain health and aging.

However, the goal is to make it both possible and practical for providers to find cognitive issues earlier, understand better the specifics and implications of the issue, and help their patients take action as soon as possible – not to replace providers at any stage.

At Linus Health we leverage AI to augment clinical expertise by surfacing signs of cognitive impairment that could otherwise be imperceptible to the human eye, as noted in your question. More specifically, for example, while a patient is taking the widely known Clock Drawing Test – now digitized – AI analyzes dozens of metrics during the person's process of drawing the clock to pick up on meaningful cues and patterns of performance.

These include subtle patterns across a wide range of insightful metrics – related to drawing efficiency, information processing, simple and complex motor components, spatial reasoning, and more – that a human wouldn't be able to catch without technology.

These metrics can be combined with other inputs like health history, life habits, symptoms and social determinants of health and run through a model grounded in specialist expertise to afford providers validated and actionable insights about the patient. In this case, AI empowers providers with new visibility into their patients' brain health so they can, in turn, empower their patients with education and next steps as appropriate.

Q. You have talked about what you perceive as the western world's bias toward drugs. What are some examples of nondrug interventions that research has proven to be largely effective?

A. First off, there's no debate around the importance of finding drugs to treat and ideally prevent Alzheimer's and other brain disorders. The concern is a concurrent, not-uncommon misconception that it's not worth detecting issues early if there's "nothing we can do'' yet.

That notion stems from the lack of drug treatments available today. Beyond the fact that more than 80% of older Americans say they would prefer to know early if they have signs of Alzheimer's, the stance that lack of drugs means lack of possibility simply doesn't reflect the full picture of options available to people at present.

Encouragingly, there is a widening body of scientific evidence supporting the impact that health and lifestyle-related changes can make on warding off and/or reducing the impact of brain health issues. The Lancet Commission's list of 12 modifiable dementia risk factors is a key example, but there are numerous other studies with similar findings, such as the FINGER Study.

The 12 in the Lancet report include a mix of health (for example, diabetes, depression, hypertension) and lifestyle-related (for example, inactivity, infrequent social contact) factors, alongside other types of factors (for example, air pollution, education level).

There also are various other studies that have focused on and supported the potential impact these types of adjustments can make on slowing cognitive decline, such as improving frailty, exercising regularly and following certain diets. This list continues to grow as researchers perform new studies to uncover more ways for people to have greater control over their brain health.

Organizations across the world are working hard to discover groundbreaking drug treatments for diseases like Alzheimer's, but there is no time to wait to address brain health issues in the meantime. While this work is ongoing, we need to empower providers and patients with novel brain health insights and evidence-based health and lifestyle interventions that have the potential to meaningfully improve patients' and their families' lives.

The sooner people know about the impact such changes could make on their brain health, the better. Also, it is unlikely that a silver bullet exists to treat brain-related disorders. The brain is incredibly complex. Neurocognitive health assessment must be multifaceted, and interventions must be multipronged and personal, which together necessitates the use of digital tools to adequately equip health systems to meet this growing need.

Twitter: @SiwickiHealthIT
Email the writer: bsiwicki@himss.org
Healthcare IT News is a HIMSS Media publication.

Want to get more stories like this one? Get daily news updates from Healthcare IT News.
Your subscription has been saved.
Something went wrong. Please try again.