An Evaluation of a Wrist Motion Tracking Algorithm to Detect Eating Activities on 408 People

Surya Sharma1, Phillip Jasper, Eric Muth, Adam Hoover

  • 1Clemson University

Details

18:15 - 20:15 | Mon 5 Mar | Caribbean ABC | MoPO.58

Session: Poster Session # 1 and BSN Innovative Health Technology Demonstrations

Full Text

Digital Library Website

Abstract

Our research considers the problem of detecting eating by tracking wrist movements. An eating activity (EA) is any meal or snack. Previously, our group developed an algorithm which reported 81% accuracy in detecting eating activities for 43 people across 449 hours of recorded activity. To further this research, we have collected 4,680 hours of activity data from 408 participants.