Objective Assessment of Strength Training Exercises using a Wrist-Worn Accelerometer

The 2008 Physical Activity Guidelines for Americans recommends that all adults perform muscle-strengthening exercises to work all of the major muscle groups of the body on at least two days per week, in addition to aerobic activity. Studies using objective methods of monitoring physical activity hav...

Full description

Saved in:
Bibliographic Details
Published in:Medicine and science in sports and exercise 2016-09, Vol.48 (9), p.1847-1855
Main Authors: Conger, Scott A, Guo, Jun, Fulkerson, Scott M, Pedigo, Lauren, Chen, Hao, Bassett, David R
Format: Article
Language:English
Subjects:
Accelerometry - instrumentation
Adult
Algorithms
Female
Humans
Male
Resistance Training
Support Vector Machine
Wrist
Young Adult
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The 2008 Physical Activity Guidelines for Americans recommends that all adults perform muscle-strengthening exercises to work all of the major muscle groups of the body on at least two days per week, in addition to aerobic activity. Studies using objective methods of monitoring physical activity have focused primarily on the assessment of aerobic activity. To date, a method for assessing resistance training (RT) exercises has not been developed using a wrist worn activity monitor. PURPOSEThe purpose of this study was to examine the use of a wrist-worn, tri-axial accelerometer-based activity monitor for classifying upper- and lower-body dumbbell RT exercises. METHODSSixty participants performed 10 repetitions each of 12 different upper and lower body dynamic dumbbell exercises. Algorithms for classifying the exercises were developed using two different methodssupport vector machine (SVM) and cosine similarity. Confusion matrices were developed for each method and inter-method reliability were assessed using Cohen’s kappa. A repeated measures ANOVA was used to compare the predicted repetitions, identified from the largest acceleration peaks, with the actual repetitions. RESULTSThe results indicated that SVM and cosine similarity accurately classified the 12 different RT exercises 78% and 85% of the time, respectively. Both methods struggled to correctly differentiate bench press vs shoulder press and squat vs walking lunges. Repetition estimates were not significantly different for 8 of the 12 exercises. For the four exercises that were significantly different, the differences amounts to less than 10%. CONCLUSIONThis study demonstrated that RT exercises can be accurately classified using a single activity monitor worn on the wrist.
ISSN:0195-9131
1530-0315
DOI:10.1249/MSS.0000000000000949