AI Can Detect Early Signs of Alzheimer’s in How People Speak, New Study Shows

UT Southwestern's Ihab Hajjar, M.D., led the study, working with a research team at Atlanta's Emory University. After being given standard cognitive assessments, 206 participants were asked to talk about an art work. Here's what happened next.

Subtle changes in the way people speak could help doctors diagnose cognitive impairment and Alzheimer’s disease—even before symptoms begin to show. 

That’s according to a new study led a UT Southwestern researcher, working with a research team at Emory University in Atlanta.

“Our focus was on identifying subtle language and audio changes that are present in the very early stages of Alzheimer’s disease but not easily recognizable by family members or an individual’s primary care physician,” Ihab Hajjar, M.D., professor of neurology at UT Southwestern’s Peter O’Donnell Jr. Brain Institute, said in a statement.

Study used advanced machine learning and NLP tools

Ihab Hajjar M.D. professor of neurology at UT Southwestern’s Peter O’Donnell Jr. Brain Institute [Photo: UTSW]

The team used advanced machine learning and natural language processing (NLP) tools to assess speech patterns in 206 people enrolled in the Emory study. Of those, 114 met the criteria for mild cognitive decline, while 92 were unimpaired. The team then mapped their findings to known biomarkers to determine their efficacy in measuring impairment.

Participants were asked to talk about a work of art

Before testing the study participants’ speech, the team first gave them standard cognitive assessments. Then each participant was asked to record “a spontaneous 1- to 2-minute description of artwork.”

“The recorded descriptions of the picture provided us with an approximation of conversational abilities that we could study via artificial intelligence to determine speech motor control, idea density, grammatical complexity, and other speech features,” Dr. Hajjar said.

Cerebral spinal fluid samples, MRI scans

Going far deeper than merely analyzing voices, the team compared the participants’ speech analytics to their cerebral spinal fluid samples and MRI scans. This was done to determine how accurate the “digital voice biomarkers” were at detecting both mild cognitive impairment and Alzheimer’s disease status and progression.

“Prior to the development of machine learning and NLP, the detailed study of speech patterns in patients was extremely labor intensive and often not successful because the changes in the early stages are frequently undetectable to the human ear,” Hajjar said in the statement.

“This novel method of testing performed well in detecting those with mild cognitive impairment and more specifically in identifying patients with evidence of Alzheimer’s disease—even when it cannot be easily detected using standard cognitive assessments,” he added.

Could save time in detecting Alzheimer’s

Traditional neuropsychological tests usually take several hours to conduct. But during this study, researchers spent under 10 minutes capturing a patient’s voice recording—pointing the way toward a more streamlined and flexible diagnostic solution.

“If confirmed with larger studies, the use of artificial intelligence and machine learning to study vocal recordings could provide primary care providers with an easy-to-perform screening tool for at-risk individuals,” Hajjar said. “Earlier diagnoses would give patients and families more time to plan for the future and give clinicians greater flexibility in recommending promising lifestyle interventions.”

Hajjar led a unit at Emory before joining UTSW in 2022

Hajjar’s work with the Emory research team reflected his own pre-UTSW history. He previously served as director of the clinical trial unit at Emory’s Goizueta Alzheimer’s Disease Research Center before joining UT Southwestern in 2022.

Hajjar is continuing to gather voice recordings in Dallas as part of a UT Southwestern follow-up study that’s being funded with a National Institutes of Health grant.

This study was supported by grants from the National Institutes of Health/National Institute on Aging and the Alzheimer’s Drug Discovery Foundation.

At UT Southwestern, Hajjar holds the Pogue Family Distinguished University Chair in Alzheimer’s Disease Clinical Research and Care, in memory of Maurine and David Weigers McMullan.

The study described in this article was published in the Alzheimer’s Association publication Diagnosis, Assessment & Disease Monitoring.

Get on the list.
Dallas Innovates, every day.

Sign up to keep your eye on what’s new and next in Dallas-Fort Worth, every day.

One quick signup, and you’re done.

R E A D   N E X T

  • Slated to be built in Fort Worth's Historic Southside neighborhood, the planned $70 million museum will get the city funding once the balance for the project has been raised. Designed by the New York office of Denmark-based Bjarke Ingels Group, the building will house the museum on its second level, with a business incubator, restaurant, 250-seat amphitheater, and storefronts at ground level. “Literally and figuratively, it was designed to be a beacon of light in an area that has been dark for a very long time,” says Jarred Howard, principal of the project's developer.

  • On Rare Disease Day, Plano Biotech Reata Pharmaceuticals Gets Its First FDA Approval

    The FDA approval is not only a first for Reata's treatment of ultra-rare disease Friedreich's Ataxia. It's a first in the company's history, reports FiercePharma. On the news, Reata’s stock skyrocketed nearly 200% Wednesday, according to the Dallas Morning News, which cited an estimate that the drug will create a revenue stream in excess of $1 billion.

  • The foundation—launched in the early 1990s by Fort Worth billionaire and philanthropist Richard E. Rainwater—said that this year's $400K Outstanding Innovation Prize will be awarded to a team from the University of California San Diego and Washington University. Dr. Susanne Wegmann of the German Center for Neurodegenerative Diseases (DZNE) will be awarded the $200K Rainwater Prize for Innovative Early-Career Scientist.

  • The nation’s highest honor for technological achievement will be bestowed by President Biden, the USPTO said. The medal is awarded to individuals, teams, and companies for outstanding contributions to the nation’s economic, environmental, and social well-being—through tech products, tech innovation, and strengthening America's tech workforce.

  • As New Year's Eve approaches, thoughts are turning to holidays, fireworks, and—yes—business plans for a whole new year. That may be why December is National Business Plan Month, and TWU's Tracy Irby has the perfect way to celebrate it—by avoiding 5 key business plan mistakes.