Segmenting accelerometer data from daily life with unsupervised machine learning.
<h4>Purpose</h4>Accelerometers are increasingly used to obtain valuable descriptors of physical activity for health research. The cut-points approach to segment accelerometer data is widely used in physical activity research but requires resource expensive calibration studies and does no...
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Public Library of Science (PLoS)
2019-01-01
|
Series: | PLoS ONE |
Online Access: | https://doi.org/10.1371/journal.pone.0208692 |
id |
doaj-678df587ea114bd4a12e5bdccf291d7b |
---|---|
record_format |
Article |
spelling |
doaj-678df587ea114bd4a12e5bdccf291d7b2021-03-04T12:39:10ZengPublic Library of Science (PLoS)PLoS ONE1932-62032019-01-01141e020869210.1371/journal.pone.0208692Segmenting accelerometer data from daily life with unsupervised machine learning.Dafne van KuppeveltJoe HeywoodMark HamerSéverine SabiaEmla FitzsimonsVincent van Hees<h4>Purpose</h4>Accelerometers are increasingly used to obtain valuable descriptors of physical activity for health research. The cut-points approach to segment accelerometer data is widely used in physical activity research but requires resource expensive calibration studies and does not make it easy to explore the information that can be gained for a variety of raw data metrics. To address these limitations, we present a data-driven approach for segmenting and clustering the accelerometer data using unsupervised machine learning.<h4>Methods</h4>The data used came from five hundred fourteen-year-old participants from the Millennium cohort study who wore an accelerometer (GENEActiv) on their wrist on one weekday and one weekend day. A Hidden Semi-Markov Model (HSMM), configured to identify a maximum of ten behavioral states from five second averaged acceleration with and without addition of x, y, and z-angles, was used for segmenting and clustering of the data. A cut-points approach was used as comparison.<h4>Results</h4>Time spent in behavioral states with or without angle metrics constituted eight and five principal components to reach 95% explained variance, respectively; in comparison four components were identified with the cut-points approach. In the HSMM with acceleration and angle as input, the distributions for acceleration in the states showed similar groupings as the cut-points categories, while more variety was seen in the distribution of angles.<h4>Conclusion</h4>Our unsupervised classification approach learns a construct of human behavior based on the data it observes, without the need for resource expensive calibration studies, has the ability to combine multiple data metrics, and offers a higher dimensional description of physical behavior. States are interpretable from the distributions of observations and by their duration.https://doi.org/10.1371/journal.pone.0208692 |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Dafne van Kuppevelt Joe Heywood Mark Hamer Séverine Sabia Emla Fitzsimons Vincent van Hees |
spellingShingle |
Dafne van Kuppevelt Joe Heywood Mark Hamer Séverine Sabia Emla Fitzsimons Vincent van Hees Segmenting accelerometer data from daily life with unsupervised machine learning. PLoS ONE |
author_facet |
Dafne van Kuppevelt Joe Heywood Mark Hamer Séverine Sabia Emla Fitzsimons Vincent van Hees |
author_sort |
Dafne van Kuppevelt |
title |
Segmenting accelerometer data from daily life with unsupervised machine learning. |
title_short |
Segmenting accelerometer data from daily life with unsupervised machine learning. |
title_full |
Segmenting accelerometer data from daily life with unsupervised machine learning. |
title_fullStr |
Segmenting accelerometer data from daily life with unsupervised machine learning. |
title_full_unstemmed |
Segmenting accelerometer data from daily life with unsupervised machine learning. |
title_sort |
segmenting accelerometer data from daily life with unsupervised machine learning. |
publisher |
Public Library of Science (PLoS) |
series |
PLoS ONE |
issn |
1932-6203 |
publishDate |
2019-01-01 |
description |
<h4>Purpose</h4>Accelerometers are increasingly used to obtain valuable descriptors of physical activity for health research. The cut-points approach to segment accelerometer data is widely used in physical activity research but requires resource expensive calibration studies and does not make it easy to explore the information that can be gained for a variety of raw data metrics. To address these limitations, we present a data-driven approach for segmenting and clustering the accelerometer data using unsupervised machine learning.<h4>Methods</h4>The data used came from five hundred fourteen-year-old participants from the Millennium cohort study who wore an accelerometer (GENEActiv) on their wrist on one weekday and one weekend day. A Hidden Semi-Markov Model (HSMM), configured to identify a maximum of ten behavioral states from five second averaged acceleration with and without addition of x, y, and z-angles, was used for segmenting and clustering of the data. A cut-points approach was used as comparison.<h4>Results</h4>Time spent in behavioral states with or without angle metrics constituted eight and five principal components to reach 95% explained variance, respectively; in comparison four components were identified with the cut-points approach. In the HSMM with acceleration and angle as input, the distributions for acceleration in the states showed similar groupings as the cut-points categories, while more variety was seen in the distribution of angles.<h4>Conclusion</h4>Our unsupervised classification approach learns a construct of human behavior based on the data it observes, without the need for resource expensive calibration studies, has the ability to combine multiple data metrics, and offers a higher dimensional description of physical behavior. States are interpretable from the distributions of observations and by their duration. |
url |
https://doi.org/10.1371/journal.pone.0208692 |
work_keys_str_mv |
AT dafnevankuppevelt segmentingaccelerometerdatafromdailylifewithunsupervisedmachinelearning AT joeheywood segmentingaccelerometerdatafromdailylifewithunsupervisedmachinelearning AT markhamer segmentingaccelerometerdatafromdailylifewithunsupervisedmachinelearning AT severinesabia segmentingaccelerometerdatafromdailylifewithunsupervisedmachinelearning AT emlafitzsimons segmentingaccelerometerdatafromdailylifewithunsupervisedmachinelearning AT vincentvanhees segmentingaccelerometerdatafromdailylifewithunsupervisedmachinelearning |
_version_ |
1714802025568403456 |