The effectiveness of eye-tracking technology in identifying people who have a genetic tendency to Alzheimer’s Disease, years before their symptoms show, has been highlighted in new research.
Tests which used the ViewMind Atlas™ system had high levels of accuracy in identifying people who had an inherited form of the disease caused by a genetic mutation and were in the symptomatic stages, and those who were asymptomatic but carried a genetic mutation, which inevitably leads to development of the condition.
ViewMind Atlas™ integrates eye-tracking and Software as a Medical Device (SaMD) to provide functional analysis of brain health.
Subtle changes
This approach allows for the detection of subtle cognitive changes that may reflect disease progression or measure a patient's response to treatment over time. The inclusion of ViewMind Atlas™ provides a distinctive opportunity to further validate non-invasive methods for early Alzheimer’s detection.
ViewMind Atlas™ was used to detect carriers of the mutation and was 100% accurate for those who were already displaying symptoms, with 96% for those who were asymptomatic.
The research, carried out in collaboration with the University of Strathclyde, confirmed that this novel methodology is more accurate than traditional cognitive tests, which often do not detect Alzheimer’s until its symptoms become more apparent. Participants in the research were drawn from extended families in Colombia in which the mutation is carried. The study, published in the journal Brain Communications, could offer a potential biomarker solution for preclinical detection of Alzheimer's disease.
Professor Mario Parra Rodriguez, Head of Strathclyde’s Department of Psychological Sciences and Health, and the lead author, said: “Most of the diagnostic approaches used in dementia are expensive and invasive, because they require injection of chemical radiotracers or extraction of fluids from the body that normally require hospital settings.
ViewMind Atlas™ is helping us to predict Alzheimer’s dementia in people years before it becomes symptomatic. Doctors will receive people who are probably not yet at the stage of dementia, but they are noticing that something is going on and may have a family history.
“One of the greatest puzzles that providers face is to decide what the potential problem could be; is it age-related forgetfulness and not necessarily dementia? Or do these cognitive problems point towards a risk of dementia in the future, so that they could act promptly rather than waiting until the person is symptomatic?”
Participants in the research undertook the Visual Short-Term Memory Binding Task, in which they assessed and compared various groups of objects, according to colour and shape. The eye-tracking AI framework used Random Forest, a supervised machine learning algorithm, to help distinguish between groups of people according to the presence or possibility of Alzheimer’s Disease.
The research was conducted in partnership with Universidad de Antioquia in Medellín, Colombia, and with ViewMind.