Assessment of wear/nonwear time classification algorithms for triaxial accelerometer - PubMed (original) (raw)

Assessment of wear/nonwear time classification algorithms for triaxial accelerometer

Leena Choi et al. Med Sci Sports Exerc. 2012 Oct.

Abstract

Purpose: The objective of this study is to assess the performance of existing wear/nonwear time classification algorithms for accelerometry data collected in the free-living environment using a wrist-worn triaxial accelerometer and a waist-worn uniaxial accelerometer in older adults.

Methods: Twenty-nine adults age 76 to 96 yr wore wrist accelerometers for approximately 24 h per day and waist accelerometers during waking for approximately 7 d of free living. Wear and nonwear times were classified by existing algorithms (Alg([ActiLife]), Alg([Troiano]), and Alg([Choi])) and compared with wear and nonwear times identified by data plots and diary records. With the use of bias and probability of correct classification, the performance of the algorithms, two time windows (60 and 90 min), and vector magnitude (VM) versus vertical axis (V) counts from a triaxial accelerometer were compared.

Results: Automated algorithms (Alg([Choi]) and Alg([Troiano])) classified wear/nonwear time intervals more accurately from VM than V counts. The use of the 90-min time window improved wear/nonwear classification accuracy when compared with the 60-min window. The Alg([Choi]) and Alg([Troiano]) performed better than the manufacturer-provided algorithm (Alg([ActiLife])), and the Alg([Choi]) performed better than the Alg([Troiano]) for wear/nonwear time classification using the data collected by both accelerometers.

Conclusions: Triaxial wrist-worn accelerometer can be used for an accurate wear/nonwear time classification in free-living older adults. The use of the 90-min window and the VM counts improves the performance of commonly used algorithms for wear/nonwear classification for both uniaxial and triaxial accelerometers.

PubMed Disclaimer

Conflict of interest statement

Conflict of Interest: none.

Figures

Figure 1

Figure 1

(A) and (B) The first two days of representative data for two subjects who wore wrist monitor for 24-h and waist monitor during waking period: wrist VM (top), wrist V (middle), and waist V (bottom). Wear and nonwear identification during period classified as sleep are represented by dashed arrows. Intervals identified as nonwear during waking are represented by solid arrows in (B).

Figure 1

Figure 1

(A) and (B) The first two days of representative data for two subjects who wore wrist monitor for 24-h and waist monitor during waking period: wrist VM (top), wrist V (middle), and waist V (bottom). Wear and nonwear identification during period classified as sleep are represented by dashed arrows. Intervals identified as nonwear during waking are represented by solid arrows in (B).

Figure 2

Figure 2

Box plots of individual bias in the daily average wear time (identified - classified) (min) classified by algorithms with two window time settings (60- and 90-min) using wrist (VM and V) and waist (V) counts data during 24-h and both monitors-worn waking periods: (A) Alg[Troiano]; (B) Alg[Choi]. The solid circles represent individual data points (N=29).

Figure 2

Figure 2

Box plots of individual bias in the daily average wear time (identified - classified) (min) classified by algorithms with two window time settings (60- and 90-min) using wrist (VM and V) and waist (V) counts data during 24-h and both monitors-worn waking periods: (A) Alg[Troiano]; (B) Alg[Choi]. The solid circles represent individual data points (N=29).

Figure 3

Figure 3

Box plots of individual probability of correct classification by algorithms with two window time settings (60- and 90-min) using wrist (VM and V) and waist (V) counts data during 24-h and both monitors-worn waking periods: (A) Alg[Troiano]; (B) Alg[Choi]. The solid circles represent individual data points (N=29).

Figure 3

Figure 3

Box plots of individual probability of correct classification by algorithms with two window time settings (60- and 90-min) using wrist (VM and V) and waist (V) counts data during 24-h and both monitors-worn waking periods: (A) Alg[Troiano]; (B) Alg[Choi]. The solid circles represent individual data points (N=29).

Similar articles

Cited by

References

    1. Bassett DR, Jr, Rowlands A, Trost SG. Calibration and validation of wearable monitors. Med Sci Sports Exerc. 2012;44(1 Suppl 1):S32–8. - PMC - PubMed
    1. Buchowski MS, Acra S, Majchrzak KM, Sun M, Chen KY. Patterns of physical activity in free-living adults in the southern united states. Eur J Clin Nutr. 2007;58(5):828–37. - PubMed
    1. Choi L, Liu Z, Matthews CE, Buchowski MS. Physicalactivity: Process physical activity accelerometer data. 2010:1–14.
    1. Choi L, Liu Z, Matthews CE, Buchowski MS. Validation of accelerometer wear and nonwear time classification algorithm. Med Sci Sports Exerc. 2011;43(2):357–64. - PMC - PubMed
    1. Freedson P, Bowles HR, Troiano R, Haskell W. Assessment of physical activity using wearable monitors: Recommendations for monitor calibration and use in the field. Med Sci Sports Exerc. 2012;44(1 Suppl 1):S1–4. - PMC - PubMed

Publication types

MeSH terms

Grants and funding

LinkOut - more resources