A new artificial intelligence-powered tool aims to detect signs of dementia, Alzheimer’s and other memory disorders by analyzing a person’s speech and language patterns.
The system is called CognoSpeak. Researchers at the University of Sheffield in the U.K. developed it.
In early trials that included both Alzheimer’s patients and cognitively heathy people, the tool showed 90% accuracy in identifying those with dementia — which is just as accurate as “pen-and-paper tests,” according to a press release announcing the new tool.
Researchers next plan to test the tool on 700 participants from U.K. memory clinics, with funding provided by the National Institute for Health and Care Research.
The goal is for CognoSpeak to help enable faster diagnosis by serving as a “middle man” between the general practitioner and the dementia specialist.
“This tool could help patients start treatments sooner, reduce waiting times and give people certainty earlier.”
“Waiting for a possible diagnosis of dementia can be a very anxious time for patients and their families,” said Dr. Dan Blackburn from the University of Sheffield’s Department of Neuroscience in the press release. “This tool could help patients start treatments sooner, reduce waiting times and give people certainty earlier.”
He added, “This would also free up clinicians’ valuable time and mean that those who need specialist care get access to it as quickly as possible.”
Dr. Michael Kleiman, PhD, a research assistant professor of neurology at the Comprehensive Center for Brain Health at the University of Miami Miller School of Medicine, is a colleague of Galvin’s who has studied the link between speech behavior and detection of early stages of cognitive impairment.
“Speech patterns in general are highly useful at identifying cognitive impairment, and some of the components of CognoSpeak, including picture descriptions, are frequently used by neuropsychologists and neurologists to assess and diagnose dementia,” Kleiman said. (He also has no affiliation with CognoSpeak.)
‘More work is needed’
While experts agree that the role of AI in neurology shows promise, more research is needed before it can be relied upon as a primary indicator.
While speech testing is useful, experts noted that it’s not the only component necessary for diagnosing dementia. (iStock)
“Although this technology is encouraging, more work is needed in larger, more representative study populations to further validate the utility of AI tools like this one in helping to diagnose Alzheimer’s,” said Griffin.
The physician would refer the patient to CognoSpeak and the test results would be routed back to the doctor — who would then decide whether the patient needed further memory care. (iStock)
Ideally, Percy added, the availability of non-invasive and potentially less expensive approaches to early detection and diagnosis — a combination of blood tests, simple digital tests and other tools — would make cognitive assessments available to more people.
Around 5.8 million people in the U.S. have Alzheimer’s disease and related dementias, according to the Centers for Disease Control and Prevention (CDC).