Movement prediction using accelerometers in a human population

Luo Xiao, Bing He, Annemarie Koster, Paolo Caserotti, Brittney Lange-Maia, Nancy W. Glynn, Tamara B. Harris, Ciprian M Crainiceanu

Research output: Contribution to journalArticle

Abstract

We introduce statistical methods for predicting the types of human activity at sub-second resolution using triaxial accelerometry data. The major innovation is that we use labeled activity data from some subjects to predict the activity labels of other subjects. To achieve this, we normalize the data across subjects by matching the standing up and lying down portions of triaxial accelerometry data. This is necessary to account for differences between the variability in the position of the device relative to gravity, which are induced by body shape and size as well as by the ambiguous definition of device placement. We also normalize the data at the device level to ensure that the magnitude of the signal at rest is similar across devices. After normalization we use overlapping movelets (segments of triaxial accelerometry time series) extracted from some of the subjects to predict the movement type of the other subjects. The problem was motivated by and is applied to a laboratory study of 20 older participants who performed different activities while wearing accelerometers at the hip. Prediction results based on other people's labeled dictionaries of activity performed almost as well as those obtained using their own labeled dictionaries. These findings indicate that prediction of activity types for data collected during natural activities of daily living may actually be possible.

Original languageEnglish (US)
Pages (from-to)513-524
Number of pages12
JournalBiometrics
Volume72
Issue number2
DOIs
StatePublished - Jun 1 2016

Fingerprint

accelerometry
Accelerometer
Accelerometry
Glossaries
Accelerometers
human population
Equipment and Supplies
prediction
Prediction
Population
Labels
Time series
Statistical methods
Gravitation
Innovation
Normalize
anthropometric measurements
hips
gravity
time series analysis

Keywords

  • Accelerometer
  • Activity type
  • Movelets
  • Prediction

ASJC Scopus subject areas

  • Statistics and Probability
  • Medicine(all)
  • Immunology and Microbiology(all)
  • Biochemistry, Genetics and Molecular Biology(all)
  • Agricultural and Biological Sciences(all)
  • Applied Mathematics

Cite this

Xiao, L., He, B., Koster, A., Caserotti, P., Lange-Maia, B., Glynn, N. W., ... Crainiceanu, C. M. (2016). Movement prediction using accelerometers in a human population. Biometrics, 72(2), 513-524. https://doi.org/10.1111/biom.12382

Movement prediction using accelerometers in a human population. / Xiao, Luo; He, Bing; Koster, Annemarie; Caserotti, Paolo; Lange-Maia, Brittney; Glynn, Nancy W.; Harris, Tamara B.; Crainiceanu, Ciprian M.

In: Biometrics, Vol. 72, No. 2, 01.06.2016, p. 513-524.

Research output: Contribution to journalArticle

Xiao, L, He, B, Koster, A, Caserotti, P, Lange-Maia, B, Glynn, NW, Harris, TB & Crainiceanu, CM 2016, 'Movement prediction using accelerometers in a human population', Biometrics, vol. 72, no. 2, pp. 513-524. https://doi.org/10.1111/biom.12382
Xiao L, He B, Koster A, Caserotti P, Lange-Maia B, Glynn NW et al. Movement prediction using accelerometers in a human population. Biometrics. 2016 Jun 1;72(2):513-524. https://doi.org/10.1111/biom.12382
Xiao, Luo ; He, Bing ; Koster, Annemarie ; Caserotti, Paolo ; Lange-Maia, Brittney ; Glynn, Nancy W. ; Harris, Tamara B. ; Crainiceanu, Ciprian M. / Movement prediction using accelerometers in a human population. In: Biometrics. 2016 ; Vol. 72, No. 2. pp. 513-524.
@article{45722ee0df1e4346a3dca4cd955bce49,
title = "Movement prediction using accelerometers in a human population",
abstract = "We introduce statistical methods for predicting the types of human activity at sub-second resolution using triaxial accelerometry data. The major innovation is that we use labeled activity data from some subjects to predict the activity labels of other subjects. To achieve this, we normalize the data across subjects by matching the standing up and lying down portions of triaxial accelerometry data. This is necessary to account for differences between the variability in the position of the device relative to gravity, which are induced by body shape and size as well as by the ambiguous definition of device placement. We also normalize the data at the device level to ensure that the magnitude of the signal at rest is similar across devices. After normalization we use overlapping movelets (segments of triaxial accelerometry time series) extracted from some of the subjects to predict the movement type of the other subjects. The problem was motivated by and is applied to a laboratory study of 20 older participants who performed different activities while wearing accelerometers at the hip. Prediction results based on other people's labeled dictionaries of activity performed almost as well as those obtained using their own labeled dictionaries. These findings indicate that prediction of activity types for data collected during natural activities of daily living may actually be possible.",
keywords = "Accelerometer, Activity type, Movelets, Prediction",
author = "Luo Xiao and Bing He and Annemarie Koster and Paolo Caserotti and Brittney Lange-Maia and Glynn, {Nancy W.} and Harris, {Tamara B.} and Crainiceanu, {Ciprian M}",
year = "2016",
month = "6",
day = "1",
doi = "10.1111/biom.12382",
language = "English (US)",
volume = "72",
pages = "513--524",
journal = "Biometrics",
issn = "0006-341X",
publisher = "Wiley-Blackwell",
number = "2",

}

TY - JOUR

T1 - Movement prediction using accelerometers in a human population

AU - Xiao, Luo

AU - He, Bing

AU - Koster, Annemarie

AU - Caserotti, Paolo

AU - Lange-Maia, Brittney

AU - Glynn, Nancy W.

AU - Harris, Tamara B.

AU - Crainiceanu, Ciprian M

PY - 2016/6/1

Y1 - 2016/6/1

N2 - We introduce statistical methods for predicting the types of human activity at sub-second resolution using triaxial accelerometry data. The major innovation is that we use labeled activity data from some subjects to predict the activity labels of other subjects. To achieve this, we normalize the data across subjects by matching the standing up and lying down portions of triaxial accelerometry data. This is necessary to account for differences between the variability in the position of the device relative to gravity, which are induced by body shape and size as well as by the ambiguous definition of device placement. We also normalize the data at the device level to ensure that the magnitude of the signal at rest is similar across devices. After normalization we use overlapping movelets (segments of triaxial accelerometry time series) extracted from some of the subjects to predict the movement type of the other subjects. The problem was motivated by and is applied to a laboratory study of 20 older participants who performed different activities while wearing accelerometers at the hip. Prediction results based on other people's labeled dictionaries of activity performed almost as well as those obtained using their own labeled dictionaries. These findings indicate that prediction of activity types for data collected during natural activities of daily living may actually be possible.

AB - We introduce statistical methods for predicting the types of human activity at sub-second resolution using triaxial accelerometry data. The major innovation is that we use labeled activity data from some subjects to predict the activity labels of other subjects. To achieve this, we normalize the data across subjects by matching the standing up and lying down portions of triaxial accelerometry data. This is necessary to account for differences between the variability in the position of the device relative to gravity, which are induced by body shape and size as well as by the ambiguous definition of device placement. We also normalize the data at the device level to ensure that the magnitude of the signal at rest is similar across devices. After normalization we use overlapping movelets (segments of triaxial accelerometry time series) extracted from some of the subjects to predict the movement type of the other subjects. The problem was motivated by and is applied to a laboratory study of 20 older participants who performed different activities while wearing accelerometers at the hip. Prediction results based on other people's labeled dictionaries of activity performed almost as well as those obtained using their own labeled dictionaries. These findings indicate that prediction of activity types for data collected during natural activities of daily living may actually be possible.

KW - Accelerometer

KW - Activity type

KW - Movelets

KW - Prediction

UR - http://www.scopus.com/inward/record.url?scp=85027957011&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85027957011&partnerID=8YFLogxK

U2 - 10.1111/biom.12382

DO - 10.1111/biom.12382

M3 - Article

C2 - 26288278

AN - SCOPUS:85027957011

VL - 72

SP - 513

EP - 524

JO - Biometrics

JF - Biometrics

SN - 0006-341X

IS - 2

ER -