For the last three years, researchers at IBM, in partnership with academia and industry, have been developing a means of identifying those at risk of dementia. What they have developed is a smartphone and tablet app that compresses a diagnostic test usually performed by a clinician.
The test involves asking a patient questions and recording their voice as they answer. Questions include asking them to repeat a sentence they just heard, name as many animals as they can in 30 seconds, describe something sad, or recall the previous morning's events.
Although clinicians would analyze the answers a patient gives, the IBM app uses the Watson artificial-intelligence platform to analyze the patient's voice, examining the tone of voice, pauses between words, signs of hesitation, and continuity of speech.
Project leader Aharon Satt says IBM chose this approach because they wanted the app to be language-agnostic. So far, the app can determine if a patient is dementia-free, showing signs of pre-dementia, or already suffering from early-stage dementia, with 85-percent accuracy.
For now, the app is only being used by clinicians and is not meant to be used to give a formal diagnosis.
From Tech Republic
View Full Article
Abstracts Copyright © 2015 Information Inc., Bethesda, Maryland, USA
No entries found