Description

Three methods were developed using one adult participant- linear regression using a waist-worn accelerometer, linear regression using a combination of waist, ankle, and wrist accelerometers, and a linear regression using a combination of waist, ankle, and wrist accelerometers in which activity type (arm- or leg-dominant) was also included in the model. The coefficient of determination (R²) was 0.855 (waist only), 0.924 (waist, ankle, wrist), and 0.952 (waist, ankle, wrist, with activity type in model).

Development/Validation

Sample: 1 male adult, 30 years of age

Setting: Laboratory

Activities: Activities of daily living, stationary, walking and running

Criterion: Cosmed K4b2 (VO₂)

Accelerometer(s): Freescale MMA7260Q on ankle, waist, and wrist

Validation approach: Comparison to criterion

Phase Designation

(What's this?)

This model is in Phase 0.

Instructions

Acceleration is filtered at 0.3-19 Hz and the integral of the absolute value of the accelerometer output was calculated using the below equation where Axi(n), Ayi(n), and Azi(n) represent ith sampled acceleration at every 0.01 s over the given time period for the x, y and z axes, respectively. The acceleration values were then used in the developed linear regression equations. While it is stated that the activity type could be automatically classified into two categories (arm-dominant and leg-dominant activities) according to the ratio of wrist to ankle acceleration, more information was not provided.

Source Information

Kim, D., & Kim, H. C. (2008). Estimation of activity energy expenditure based on activity classification using multi-site triaxial accelerometry. Electronics Letters, 44(4), 266-267.https://doi.org/10.1049/el:20082139 Link to Paper

Corresponding author: HC Kim, hckim@snu.ac.kr

Contact

Kimberly Clevenger at accelerometerrepository@gmail.com