Home → News → AI Can Spot Signs of Alzheimer's Before Your Family... → Full Text

AI Can Spot Signs of Alzheimer's Before Your Family Does

By Technology Review

March 22, 2018

[article image]

Researchers at the Massachusetts Institute of Technology's (MIT) Computer Science and Artificial Intelligence Laboratory are exploring machine learning as a tool for identifying signs of Alzheimer's disease that physicians and loved ones may otherwise overlook.

They have developed an AI-powered device that MIT professor Dina Katabi says can monitor patients while sparing them from wearing sensors and other cumbersome hardware.

The device emits a wireless radio signal that reflects off everything in a 30-foot radius, including human bodies, and every single movement causes a change in the reflected signal. These reflections are analyzed by machine-learning algorithms, and the AI is designed to note deviations from patterns that might signify sleep disturbances, repeated behavior, agitation, and other classic Alzheimer's symptoms.

In a patient diagnosed with Alzheimer's, the researchers could tell she was waking up and wandering early in the morning, and that she would pace more after certain relatives visited.

From Technology Review
View Full Article


Abstracts Copyright © 2018 Information Inc., Bethesda, Maryland, USA


No entries found