» Articles » PMID: 22776868

Estimating Energy Expenditure from Raw Accelerometry in Three Types of Locomotion

Overview
Specialty Orthopedics
Date 2012 Jul 11
PMID 22776868
Citations 7
Authors
Affiliations
Soon will be listed here.
Abstract

Purpose: Accuracy of estimating activity-related energy expenditure (AEE) from raw body acceleration may improve by using prediction equations that are specific for the type of activity. The current study aims to improve published equations by deriving an equation for overground walking and to evaluate whether overground cycling and stair walking require specific prediction equations.

Methods: Participants (91 male/95 female, 8-81 yr old) were equipped with a triaxial accelerometer (DynaPort MiniMod; McRoberts BV, The Hague, The Netherlands) on their lower back. Total energy expenditure (TEE) was measured using a mobile oxygen analyzer (MetaMax 3b; Cortex Biophysik, Leipzig, Germany). Resting energy expenditure (REE) was measured for 30 min, following which a physical activity course was completed involving walking on level ground at slow (8 min), normal (8 min), and fast speed (3 min), stair walking (3 min), and cycling (8 min). AEE was calculated as TEE - REE, expressed in both absolute (kJ·min) and relative (J·min·kg) units. Mixed linear regression analysis was used for developing regression equations for walking, stair walking, and cycling.

Results: Acceleration contributed 76% and 93% (P < 0.001) to explained variance in walking AEE for absolute and relative AEE models, respectively. Age and gender improved estimation accuracy by <1%. Applying a conservative walking equation, AEE (J·min·kg) = -40.19 + 816.11 acceleration (g) (root-mean-square error = 34.00 J·min·kg), to cycling and stair walking resulted in mean bias (95% limits of agreement) of -253 (-449, 46) and -276 (-442, 109) J·min·kg, respectively (approximately 50% bias). Acceleration added 35% and 42% to explained variance in relative AEE (J·min·kg) during cycling and stair walking, respectively; this fraction was approximately 20% for absolute AEE (kJ·min) in both activities.

Conclusion: AEE during walking can be predicted across a wide age range using raw acceleration, but activity-specific equations are needed for cycling and stair walking.

Citing Articles

Descriptive epidemiology of energy expenditure in the UK: findings from the National Diet and Nutrition Survey 2008-15.

Brage S, Lindsay T, Venables M, Wijndaele K, Westgate K, Collins D Int J Epidemiol. 2020; 49(3):1007-1021.

PMID: 32191299 PMC: 7394951. DOI: 10.1093/ije/dyaa005.


Segmenting accelerometer data from daily life with unsupervised machine learning.

van Kuppevelt D, Heywood J, Hamer M, Sabia S, Fitzsimons E, van Hees V PLoS One. 2019; 14(1):e0208692.

PMID: 30625153 PMC: 6326431. DOI: 10.1371/journal.pone.0208692.


Clinical Evaluation of the Measurement Performance of the Philips Health Watch: A Within-Person Comparative Study.

Hendrikx J, Ruijs L, Cox L, Lemmens P, Schuijers E, Goris A JMIR Mhealth Uhealth. 2017; 5(2):e10.

PMID: 28153815 PMC: 5314101. DOI: 10.2196/mhealth.6893.


Measuring moderate-intensity walking in older adults using the ActiGraph accelerometer.

Barnett A, van den Hoek D, Barnett D, Cerin E BMC Geriatr. 2016; 16(1):211.

PMID: 27931188 PMC: 5146877. DOI: 10.1186/s12877-016-0380-5.


Precipitation and Physical Activity in Older Adults: The Moderating Role of Functional Mobility and Physical Activity Intentions.

Hoppmann C, Lee J, Ziegelmann J, Graf P, Khan K, Ashe M J Gerontol B Psychol Sci Soc Sci. 2015; 72(5):792-800.

PMID: 26707498 PMC: 5926997. DOI: 10.1093/geronb/gbv107.