Sign In

Communications of the ACM

ACM TechNews

Team Uses Digital Cameras, ML to Predict Neurological Disease

View as: Print Mobile App Share:

How the system acquires data on gait changes over time.

Credit: Rachneet Kaur et al

A machine learning algorithm developed by University of Illinois Urbana-Champaign (U of I) researchers aims to improve the diagnosis of people with multiple sclerosis (MD) and Parkinson's disease (PD).

The algorithm can distinguish people with MS and PD from people without those diseases based on changes in gait over time.

The researchers used digital cameras to capture the movement of participants' hips and lower limbs as they walked on a treadmill, and as they walked while reciting every-other letter of the alphabet in order.

Said U of I's Kaur, "We looked at the body coordinates for hips, knees, ankles, the big and small toes, and the heels."

The algorithm was over 75% accurate in detecting differences between those with and without MS or PD.

From University of Illinois Urbana-Champaign
View Full Article


Abstracts Copyright © 2022 SmithBucklin, Washington, DC, USA


No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account