Can Artificial Intelligence Help Forestall the Global Dementia Epidemic?
With artificial intelligence steadily infiltrating nearly every aspect of modern life, there is a healthy amount of skepticism about the outsize role of machine learning in life in the twenty-first century. For many, concerns over the implications of technologies such as facial recognition, online tracking of individual preferences, and robots assuming roles formerly filled by humans raise understandable alarm. Fortunately, the medical sector provides a powerful example of artificial intelligence being leveraged to support rather than compete with human intelligence.
An unmatched tool for early detection of cognitive alterations
Generally speaking, our brain health remains somewhat of a black box until symptoms of cognitive decline present and our brain activity is finally monitored through advanced medical testing. While most of us visit our primary care physician at least once a year for a comprehensive review of our physical health, no such proactive mechanism currently exists with regard to our cognitive health.
This tendency to treat cognitive disorders only retroactively becomes increasingly problematic as we are beset by a global epidemic of diseases related to cognitive decline. The number of dementia cases is expected to reach 1 million in the UK and 7.1 million in the US by 2025. By the time most cases of Alzheimer’s disease and other forms of dementia are diagnosed, the damage is often irreversible. As researchers are discovering through longitudinal research, artificial intelligence—in particular, the analysis of digital biomarkers—can detect the signs of cognitive decline more than ten years before dementia symptoms begin to appear. Such developments create great hope for being able to create methods of detecting specific neurocognitive features of dementia years before such pathological changes become detectable via traditional forms of assessment or lead to concerns in affected individuals.
This discovery has wide-reaching implications for behavioral and pharmacological cognitive health interventions.
Digital biomarkers light the way
One AI-driven venture working to transform dementia diagnosis and outlook is ViewMind, a digital health company that develops precision diagnostics of neurocognitive disorders. We confirmed through multi-year studies that particular eye movement patterns can not only reveal the presence of cognitive decline but also predict which individuals experiencing such decline will later develop serious forms of dementia such as Alzheimer’s disease. By administering a 20-minute test that monitors eye response to visual stimuli through a virtual reality headset, we were able to predict with 98% accuracy which response patterns would correlate to a subsequent diagnosis of Alzheimer’s disease.
Such analysis of digital biomarkers allows artificial intelligence researchers to penetrate the black box of our cognition and shed light on its inner workings. The ability to detect at an early stage eye movements that correspond to the accumulation of Beta-amyloid or tau protein in brain cells, key biomarkers of Alzheimer’s disease combined with precisely measured cognitive functions, has the potential to transform the future of Alzheimer’s intervention.
Armed with the ability to detect the signs of Alzheimer’s and other forms of dementia at an early stage through such methods, researchers will be able to validate the performance of various treatments when they might actually be effective at slowing or altogether preventing the disease. These could include drug-based treatments targeting different mechanisms believed to be causing the cognitive decline, or any number of behavioral or pharmacological interventions yet to be validated.
Disease management of neurological diseases
Treating dementia can be very challenging, as the physician lacks an accurate picture of how a drug is impacting the patient. The portfolio of solutions today such as MRI, CSF (lumbar puncture) and PET are expensive and invasive and are therefore only prescribed and reimbursed at the later stages of neurodegeneration.
Neurocognitive assessment testing based on computers lacks the necessary sensitivity to monitor the impact of treatment in short timescales. Digital biomarker research, in contrast, holds the promise to create solutions that are so sensitive, the impact of treatment can be seen in a few months. Physicians can then enable physicians to manage neurological disease with objective data related to a particular patient’s response.
The Early Detection of Neurodegenerative Diseases (EdoN) initiative was launched by Alzheimer’s Research UK to further such AI-driven early detection efforts. The project will collect a large body of digital biomarker data from research subjects using smartphone apps and wearable devices, which will then be linked to clinical tests for analysis by digital technology and neurodegeneration experts across the globe. The hope is that patterns will emerge that can then be used to detect signs of neurodegeneration in individuals who are not yet displaying dementia symptoms.
The application of artificial intelligence technology in the realm of dementia research holds great potential for improved outcomes in cognitive health throughout the course of human life. With noninvasive, affordable diagnostic methods that are easy to administer and scale for implementation around the world, innovators in medical AI can help normalize preventive cognitive health and turn the tide of the dementia epidemic.