No Cover Image

Journal article 1004 views 246 downloads

Investigating optimal accelerometer placement for energy expenditure prediction in children using a machine learning approach

Kelly Mackintosh Orcid Logo, A H K Montoye, K A Pfeiffer, Melitta McNarry Orcid Logo

Physiological Measurement, Volume: 37, Issue: 10, Pages: 1728 - 1740

Swansea University Authors: Kelly Mackintosh Orcid Logo, Melitta McNarry Orcid Logo

Check full text

DOI (Published version): 10.1088/0967-3334/37/10/1728

Abstract

Accurate measurement of energy expenditure (EE) is imperative for identifying and targeting health-associated implications. Whilst numerous accelerometer-based regression equations to predict EE have been developed, there remains little consensus regarding optimal accelerometer placement. Therefore,...

Full description

Published in: Physiological Measurement
ISSN: 0967-3334 1361-6579
Published: IOP Publishing 2016
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa29306
Tags: Add Tag
No Tags, Be the first to tag this record!
first_indexed 2016-07-29T18:46:20Z
last_indexed 2020-07-17T18:45:57Z
id cronfa29306
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2020-07-17T15:56:01.4580486</datestamp><bib-version>v2</bib-version><id>29306</id><entry>2016-07-29</entry><title>Investigating optimal accelerometer placement for energy expenditure prediction in children using a machine learning approach</title><swanseaauthors><author><sid>bdb20e3f31bcccf95c7bc116070c4214</sid><ORCID>0000-0003-0355-6357</ORCID><firstname>Kelly</firstname><surname>Mackintosh</surname><name>Kelly Mackintosh</name><active>true</active><ethesisStudent>false</ethesisStudent></author><author><sid>062f5697ff59f004bc8c713955988398</sid><ORCID>0000-0003-0813-7477</ORCID><firstname>Melitta</firstname><surname>McNarry</surname><name>Melitta McNarry</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2016-07-29</date><deptcode>STSC</deptcode><abstract>Accurate measurement of energy expenditure (EE) is imperative for identifying and targeting health-associated implications. Whilst numerous accelerometer-based regression equations to predict EE have been developed, there remains little consensus regarding optimal accelerometer placement. Therefore, the purpose of the present study was to validate and compare artificial neural networks (ANNs) developed from accelerometers worn on various anatomical positions, and combinations thereof, to predict EE.Twenty-seven children (15 boys; 10.8&#x2009;&#x2009;&#xB1;&#x2009;&#x2009;1.1 years) participated in an incremental treadmill test and 30&#x2009;min exergaming session wearing a portable gas analyser and nine ActiGraph GT3X+&#x2009;&#x2009;accelerometers (chest and left and right wrists, hips, knees, and ankles). Age and sex-specific resting EE equations (Schofield) were used to estimate METs from the oxygen uptake measures. Using all the data from both exergames, incremental treadmill test and the transition period in between, ANNs were created and tested separately for each accelerometer and for combinations of two or more using a leave-one-out approach to predict EE compared to measured EE. Six features (mean and variance of the three accelerometer axes) were extracted within each 15&#x2009;s window as inputs in the ANN. Correlations and root mean square error (RMSE) were calculated to evaluate prediction accuracy of each ANN, and repeated measures ANOVA was used to statistically compare accuracy of the ANNs.All single-accelerometer ANNs and combinations of two-, three-, and four-accelerometers performed equally (r&#x2009;&#x2009;=&#x2009;&#x2009;0.77&#x2013;0.82), demonstrating higher correlations than the 9-accelerometer ANN (r&#x2009;&#x2009;=&#x2009;&#x2009;0.69) or the Freedson linear regression equation (r&#x2009;&#x2009;=&#x2009;&#x2009;0.75). RMSE did not differ between single-accelerometer ANNs or combinations of two, three, or four accelerometers (1.21&#x2013;1.31 METs), demonstrating lower RMSEs than the 9-accelerometer ANN (1.46 METs) or Freedson equation (1.74 METs).These findings provide preliminary evidence that ANNs developed from single accelerometers mounted on various anatomical positions demonstrate equivalency in the accuracy to predict EE in a semi-structured setting, supporting the use of ANNs in improving EE prediction accuracy compared with linear regression.</abstract><type>Journal Article</type><journal>Physiological Measurement</journal><volume>37</volume><journalNumber>10</journalNumber><paginationStart>1728</paginationStart><paginationEnd>1740</paginationEnd><publisher>IOP Publishing</publisher><issnPrint>0967-3334</issnPrint><issnElectronic>1361-6579</issnElectronic><keywords/><publishedDay>1</publishedDay><publishedMonth>10</publishedMonth><publishedYear>2016</publishedYear><publishedDate>2016-10-01</publishedDate><doi>10.1088/0967-3334/37/10/1728</doi><url/><notes/><college>COLLEGE NANME</college><department>Sport and Exercise Sciences</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>STSC</DepartmentCode><institution>Swansea University</institution><apcterm/><lastEdited>2020-07-17T15:56:01.4580486</lastEdited><Created>2016-07-29T13:11:06.2610519</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Sport and Exercise Sciences</level></path><authors><author><firstname>Kelly</firstname><surname>Mackintosh</surname><orcid>0000-0003-0355-6357</orcid><order>1</order></author><author><firstname>A H K</firstname><surname>Montoye</surname><order>2</order></author><author><firstname>K A</firstname><surname>Pfeiffer</surname><order>3</order></author><author><firstname>Melitta</firstname><surname>McNarry</surname><orcid>0000-0003-0813-7477</orcid><order>4</order></author></authors><documents><document><filename>29306__3394__73078bcd4da34ed5850bcd582a3d1434.pdf</filename><originalFilename>mackintosh2016.pdf</originalFilename><uploaded>2016-07-29T13:23:15.5570000</uploaded><type>Output</type><contentLength>897380</contentLength><contentType>application/pdf</contentType><version>Accepted Manuscript</version><cronfaStatus>true</cronfaStatus><embargoDate>2017-09-21T00:00:00.0000000</embargoDate><copyrightCorrect>false</copyrightCorrect></document></documents><OutputDurs/></rfc1807>
spelling 2020-07-17T15:56:01.4580486 v2 29306 2016-07-29 Investigating optimal accelerometer placement for energy expenditure prediction in children using a machine learning approach bdb20e3f31bcccf95c7bc116070c4214 0000-0003-0355-6357 Kelly Mackintosh Kelly Mackintosh true false 062f5697ff59f004bc8c713955988398 0000-0003-0813-7477 Melitta McNarry Melitta McNarry true false 2016-07-29 STSC Accurate measurement of energy expenditure (EE) is imperative for identifying and targeting health-associated implications. Whilst numerous accelerometer-based regression equations to predict EE have been developed, there remains little consensus regarding optimal accelerometer placement. Therefore, the purpose of the present study was to validate and compare artificial neural networks (ANNs) developed from accelerometers worn on various anatomical positions, and combinations thereof, to predict EE.Twenty-seven children (15 boys; 10.8  ±  1.1 years) participated in an incremental treadmill test and 30 min exergaming session wearing a portable gas analyser and nine ActiGraph GT3X+  accelerometers (chest and left and right wrists, hips, knees, and ankles). Age and sex-specific resting EE equations (Schofield) were used to estimate METs from the oxygen uptake measures. Using all the data from both exergames, incremental treadmill test and the transition period in between, ANNs were created and tested separately for each accelerometer and for combinations of two or more using a leave-one-out approach to predict EE compared to measured EE. Six features (mean and variance of the three accelerometer axes) were extracted within each 15 s window as inputs in the ANN. Correlations and root mean square error (RMSE) were calculated to evaluate prediction accuracy of each ANN, and repeated measures ANOVA was used to statistically compare accuracy of the ANNs.All single-accelerometer ANNs and combinations of two-, three-, and four-accelerometers performed equally (r  =  0.77–0.82), demonstrating higher correlations than the 9-accelerometer ANN (r  =  0.69) or the Freedson linear regression equation (r  =  0.75). RMSE did not differ between single-accelerometer ANNs or combinations of two, three, or four accelerometers (1.21–1.31 METs), demonstrating lower RMSEs than the 9-accelerometer ANN (1.46 METs) or Freedson equation (1.74 METs).These findings provide preliminary evidence that ANNs developed from single accelerometers mounted on various anatomical positions demonstrate equivalency in the accuracy to predict EE in a semi-structured setting, supporting the use of ANNs in improving EE prediction accuracy compared with linear regression. Journal Article Physiological Measurement 37 10 1728 1740 IOP Publishing 0967-3334 1361-6579 1 10 2016 2016-10-01 10.1088/0967-3334/37/10/1728 COLLEGE NANME Sport and Exercise Sciences COLLEGE CODE STSC Swansea University 2020-07-17T15:56:01.4580486 2016-07-29T13:11:06.2610519 Faculty of Science and Engineering School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Sport and Exercise Sciences Kelly Mackintosh 0000-0003-0355-6357 1 A H K Montoye 2 K A Pfeiffer 3 Melitta McNarry 0000-0003-0813-7477 4 29306__3394__73078bcd4da34ed5850bcd582a3d1434.pdf mackintosh2016.pdf 2016-07-29T13:23:15.5570000 Output 897380 application/pdf Accepted Manuscript true 2017-09-21T00:00:00.0000000 false
title Investigating optimal accelerometer placement for energy expenditure prediction in children using a machine learning approach
spellingShingle Investigating optimal accelerometer placement for energy expenditure prediction in children using a machine learning approach
Kelly Mackintosh
Melitta McNarry
title_short Investigating optimal accelerometer placement for energy expenditure prediction in children using a machine learning approach
title_full Investigating optimal accelerometer placement for energy expenditure prediction in children using a machine learning approach
title_fullStr Investigating optimal accelerometer placement for energy expenditure prediction in children using a machine learning approach
title_full_unstemmed Investigating optimal accelerometer placement for energy expenditure prediction in children using a machine learning approach
title_sort Investigating optimal accelerometer placement for energy expenditure prediction in children using a machine learning approach
author_id_str_mv bdb20e3f31bcccf95c7bc116070c4214
062f5697ff59f004bc8c713955988398
author_id_fullname_str_mv bdb20e3f31bcccf95c7bc116070c4214_***_Kelly Mackintosh
062f5697ff59f004bc8c713955988398_***_Melitta McNarry
author Kelly Mackintosh
Melitta McNarry
author2 Kelly Mackintosh
A H K Montoye
K A Pfeiffer
Melitta McNarry
format Journal article
container_title Physiological Measurement
container_volume 37
container_issue 10
container_start_page 1728
publishDate 2016
institution Swansea University
issn 0967-3334
1361-6579
doi_str_mv 10.1088/0967-3334/37/10/1728
publisher IOP Publishing
college_str Faculty of Science and Engineering
hierarchytype
hierarchy_top_id facultyofscienceandengineering
hierarchy_top_title Faculty of Science and Engineering
hierarchy_parent_id facultyofscienceandengineering
hierarchy_parent_title Faculty of Science and Engineering
department_str School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Sport and Exercise Sciences{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Sport and Exercise Sciences
document_store_str 1
active_str 0
description Accurate measurement of energy expenditure (EE) is imperative for identifying and targeting health-associated implications. Whilst numerous accelerometer-based regression equations to predict EE have been developed, there remains little consensus regarding optimal accelerometer placement. Therefore, the purpose of the present study was to validate and compare artificial neural networks (ANNs) developed from accelerometers worn on various anatomical positions, and combinations thereof, to predict EE.Twenty-seven children (15 boys; 10.8  ±  1.1 years) participated in an incremental treadmill test and 30 min exergaming session wearing a portable gas analyser and nine ActiGraph GT3X+  accelerometers (chest and left and right wrists, hips, knees, and ankles). Age and sex-specific resting EE equations (Schofield) were used to estimate METs from the oxygen uptake measures. Using all the data from both exergames, incremental treadmill test and the transition period in between, ANNs were created and tested separately for each accelerometer and for combinations of two or more using a leave-one-out approach to predict EE compared to measured EE. Six features (mean and variance of the three accelerometer axes) were extracted within each 15 s window as inputs in the ANN. Correlations and root mean square error (RMSE) were calculated to evaluate prediction accuracy of each ANN, and repeated measures ANOVA was used to statistically compare accuracy of the ANNs.All single-accelerometer ANNs and combinations of two-, three-, and four-accelerometers performed equally (r  =  0.77–0.82), demonstrating higher correlations than the 9-accelerometer ANN (r  =  0.69) or the Freedson linear regression equation (r  =  0.75). RMSE did not differ between single-accelerometer ANNs or combinations of two, three, or four accelerometers (1.21–1.31 METs), demonstrating lower RMSEs than the 9-accelerometer ANN (1.46 METs) or Freedson equation (1.74 METs).These findings provide preliminary evidence that ANNs developed from single accelerometers mounted on various anatomical positions demonstrate equivalency in the accuracy to predict EE in a semi-structured setting, supporting the use of ANNs in improving EE prediction accuracy compared with linear regression.
published_date 2016-10-01T03:35:41Z
_version_ 1763751537399562240
score 11.016235

Similar Items