The Activ8 activity monitor: Validation of posture and movement classification

Herwin Horemans*, Hedwig Kooijmans, Rita van den Berg-emons, Hans Bussmann

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

13 Citations (Web of Science)
30 Downloads (Pure)

Abstract

Introduction: To set up and guide interventions with the aim to increase physical activity and lower sedentary behavior valid monitoring of physical behavior is essential. The aim of this study was to evaluate the validity of the single-unit Activ8 activity monitor to classify several body postures and movements.Methods: Twelve healthy adults performed a series of activities, representative for everyday life, according to a standard protocol. Activ8 was both worn in the trouser pocket (prescribed location) and fixated to the front of the thigh. Activities were video recorded and analyzed thereafter. Postures and movements that were analyzed were lying/sitting, standing, walking, cycling, and running.Results: The agreement between Activ8 output and video analysis was 89.7% (inter-subject range: 66.0 to 96.6%) for the pocket location and 91.9% (range 85.5 to 95.1%) for the thigh location. Sensitivity and positive predictive value scores for both locations were all above 80%, except for standing (69% or higher). Differences in classified duration of separate postures and movements were within 20% for walking, sitting and running.Conclusion: The Activ8 is a valid instrument to quantify a defined set of body postures and movements. Because of the smaller time difference, the thigh location is preferred for research purposes.
Original languageEnglish
Article number2055668319890535
Number of pages7
JournalJournal of Rehabilitation and Assistive Technologies Engineering
Volume7
DOIs
Publication statusPublished - Mar 2020

Research programs

  • EMC OR-01

Fingerprint

Dive into the research topics of 'The Activ8 activity monitor: Validation of posture and movement classification'. Together they form a unique fingerprint.

Cite this