Diagnosing Alzheimer’s disease can be a tricky and lengthy process, with multiple in-person neuropsychological exams followed by even more tests. Diagnosis could very well cost a patient valuable treatment time, so speed is of the utmost importance.
Researchers from Boston University have developed an artificial intelligence program that can automate and optimize the diagnosis process and maybe do it online.
Close listening
The new machine-learning computational model picks up on cognitive impairment from audio recordings of patients’ neuropsychological interviews. It uses a technique called natural language processing that computers use to understand text. The transcriptions of these interviews are then turned into numbered, usable data which that program interprets for signs of cognitive impairment.
“This approach brings us one step closer to early intervention,” says Ioannis Paschalidis, a coauthor on the paper and a BU College of Engineering Distinguished Professor of Engineering. “It can form the basis of an online tool that could reach everyone and could increase the number of people who get screened early.”
The tests looked at 1,000 individuals from the Framingham Heart Study. Their findings showed that the program was able to distinguish between those with dementia and mild cognitive impairment. Interestingly, the sound of the recording and the way that the patients spoke were less important to diagnosis than what they were saying.
“It surprised us that speech flow or other audio features are not that critical; you can automatically transcribe interviews reasonably well, and rely on text analysis through AI to assess cognitive impairment,” says Paschalidis.
While the team still needs to run more tests to prove the effectiveness of the program, these early findings bode well. This machine-learning program could provide clinicians and doctors a quicker method for diagnosing Alzheimer’s disease, providing valuable time for early treatment and measures to slow the disease’s progression.
“Our models can help clinicians assess patients in terms of their chances of cognitive decline,” says Paschalidis, “and then best tailor resources to them by doing further testing on those that have a higher likelihood of dementia.”
Source Study: Alzheimer’s Association — Automated detection of mild cognitive impairment and dementia from voice recordings: A natural language processing approach – Amini – – Alzheimer’s & Dementia – Wiley Online Library
