Lookup NU author(s): Dr Jan Smeddinck
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND).
The prevalence of a sedentary lifestyle is a major contributor to many chronic afflictions in modern society. Objective study and monitoring to gain an accurate understanding of situated sedentary behavior, for example when at home, present considerable challenges, e.g. regarding ecological validity. Non-intrusive monitoring based on Wi-Fi signals provides a new way to gain insights into populations that are at risk of the negative effects of a sedentary lifestyle, or who are already in functional rehabilitation. In this paper we describe a tracking technology for everyday activities that consists of two parts: (1) recognizing general physical activity, as well as the activities of common classes; and (2) measuring the statistical duration of these recognized categories. Employing common commercial Wi-Fi equipment, we performed validation studies in a typical noisy family home environment, achieving the following key results: (1) a recognition rate of the general presence of physical activity of 99.05%, an average recognition rate of 92% when detecting four common classes of activities; and (2) Kappa coefficient analysis to evaluate the consistency of the statistical duration of the automatic activity detection based on Wi-Fi signals and manually coded activity detection based on camera recordings. The coefficient for the presence of general physical activity of .93 and the average consistency coefficient of the classified activity categories of .72 suggest a high reliability of the automatic detection outcomes. This work aims to support both research and interventions for the prevention, treatment, and rehabilitation of the consequences of a sedentary lifestyle, by establishing new technologies and methods for observing everyday functional activities that are crucial for individual independent living and well-being.
Author(s): Zhang H, Smeddinck J, Malaka R, Shu Y, Chen C, He B, Fu Z, Lawo M
Publication type: Article
Publication status: Published
Journal: Pervasive and Mobile Computing
Print publication date: 01/03/2019
Online publication date: 30/01/2019
Acceptance date: 25/01/2019
Date deposited: 12/05/2020
ISSN (print): 1574-1192
Publisher: Elsevier BV
Altmetrics provided by Altmetric