Researchers from Toronto Rehabilitation Institute examined the speech samples (transcripts and audio files) of 167 patients diagnosed with “possible” or “probable” Alzheimer’s disease as well as 97 healthy controls who were asked to describe a picture during an interview. Based on a comprehensive analysis of the samples obtained, the researchers found that they could create a language software program that was able to distinguish people with Alzheimer’s disease from those without with about 82% accuracy.
The program examined 370 elements of speech, though maximum accuracy was achieved using a set of 35 speech features that primarily fell into four categories: semantic impairment (using overly simple words), acoustic impairment (speaking very slowly), syntax impairment (using less complex grammar), and information impairment (not clearly identifying the main aspects of the picture).
“Although memory impairment may be the definitive symptom for the diagnosis of AD, it is not necessarily the most sensitive index of cognitive function and response to intervention,” the study authors wrote. “Computational analyses of naturalistic language may ultimately provide a means to monitor changes in cognitive status over the course of the disease, as well as responsiveness to interventions, and can thus serve as a useful clinical tool for purposes well beyond diagnosis.”
To read about additional research into early detection and diagnosis of Alzheimer's, see the Psychiatric News article “Cardiovascular Risk Factors May Serve as Early Indicator of Cognitive Decline.”
(Image: Diego Cervo/Shutterstock)