Literature DB >> 30509863

Development of a gold-standard method for the identification of sedentary, light and moderate physical activities in older adults: Definitions for video annotation.

Alan Kevin Bourke1, Espen Alexander F Ihlen2, Jorunn Lægdheim Helbostad2.   

Abstract

OBJECTIVES: The development of a reliable method for the identification of sedentary, light and moderate physical activities in older adults. The method consists of a validated set of definitions for the identification of the initiation and termination of physical activities performed by older adult participants, video recorded during free-living and a laboratory setting.
DESIGN: Inter-rater reliability assessment in a fully crossed design.
METHODS: An iterative consensus process was used to define the initiation and termination of common activities of daily living. These definitions were then tested using videos recorded in two scenarios (1) by 9 raters who annotated a video recording, of a free-living protocol in a home environment, recorded in a first person view, using a body-worn camera and (2) by 7 raters who annotated a video recording, of older adults performing a semi-structured protocol in a living-lab environment, recorded in a third person view, using wall mounted cameras.
RESULTS: Inter-rater reliability was excellent for all items, with Krippendorff's alpha and Fleiss' kappa all above 0.84 and a percentage of agreement above 88%. All ICC(C,1) inter-rater values for the activity quantity and duration were all above 0.9.
CONCLUSIONS: This set of physical activity initiation and termination definitions offers independent researchers a gold standard method to allow for the consistent annotation of high-frequency video footage (25fps), in both a free-living and laboratory setting. When synchronised with body-worn or ambient sensors, this annotation will allow for the development and validation of physical activity classification systems to a higher resolution than before.
Copyright © 2018 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.

Entities:  

Keywords:  Human activity recognition; Physical activity definitions; Video gold standard

Mesh:

Year:  2018        PMID: 30509863     DOI: 10.1016/j.jsams.2018.11.011

Source DB:  PubMed          Journal:  J Sci Med Sport        ISSN: 1878-1861            Impact factor:   4.319


  2 in total

1.  Optimization and Validation of an Adjustable Activity Classification Algorithm for Assessment of Physical Behavior in Elderly.

Authors:  Wouter Bijnens; Jos Aarts; An Stevens; Darcy Ummels; Kenneth Meijer
Journal:  Sensors (Basel)       Date:  2019-12-04       Impact factor: 3.576

2.  Template-Based Recognition of Human Locomotion in IMU Sensor Data Using Dynamic Time Warping.

Authors:  Kim S Sczuka; Marc Schneider; Alan K Bourke; Sabato Mellone; Ngaire Kerse; Jorunn L Helbostad; Clemens Becker; Jochen Klenk
Journal:  Sensors (Basel)       Date:  2021-04-07       Impact factor: 3.576

  2 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.