No Cover Image

Journal article 1160 views 291 downloads

A machine learning approach to measure and monitor physical activity in children

Paul Fergus, Abir J. Hussain, John Hearty, Stuart Fairclough, Lynne Boddy, Kelly Mackintosh Orcid Logo, Gareth Stratton Orcid Logo, Nicky Ridgers, Dhiya Al-Jumeily, Ahmed J. Aljaaf, Jenet Lunn

Neurocomputing, Volume: 228, Pages: 220 - 230

Swansea University Authors: Kelly Mackintosh Orcid Logo, Gareth Stratton Orcid Logo

Check full text

DOI (Published version): 10.1016/j.neucom.2016.10.040

Abstract

The growing trend of obesity and overweight worldwide has reached epidemic proportions with one third of the global population now considered obese. This is having a significant medical impact on children and adults who are at risk of developing osteoarthritis, coronary heart disease and stroke, typ...

Full description

Published in: Neurocomputing
ISSN: 0925-2312
Published: 2017
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa30942
Tags: Add Tag
No Tags, Be the first to tag this record!
first_indexed 2016-11-04T20:28:31Z
last_indexed 2018-02-09T05:17:20Z
id cronfa30942
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2017-01-17T09:32:57.7310355</datestamp><bib-version>v2</bib-version><id>30942</id><entry>2016-11-04</entry><title>A machine learning approach to measure and monitor physical activity in children</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>6d62b2ed126961bed81a94a2beba8a01</sid><ORCID>0000-0001-5618-0803</ORCID><firstname>Gareth</firstname><surname>Stratton</surname><name>Gareth Stratton</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2016-11-04</date><deptcode>STSC</deptcode><abstract>The growing trend of obesity and overweight worldwide has reached epidemic proportions with one third of the global population now considered obese. This is having a significant medical impact on children and adults who are at risk of developing osteoarthritis, coronary heart disease and stroke, type 2 diabetes, cancers, respiratory problems, and non-alcoholic fatty liver disease. In an attempt to redress the issue, physical activity is being promoted as a fundamental component for maintaining a healthy lifestyle. Recommendations for physical activity levels are issued by most governments as part of their public health measures. However, current techniques and protocols, including those used in laboratory settings, have been criticised. The main concern is that it is not feasible to use multiple pieces of measurement hardware, such as VO2 masks and heart rate monitors, to monitor children in free-living environments due to weight and encumbrance constraints. This has prompted research in the use of wearable sensing and machine learning technology to produce classifications for specific physical activity events. This paper builds on this approach and presents a supervised machine learning method that utilises data obtained from accelerometer sensors worn by children in free-living environments. Our results show that when using an artificial neural network algorithm it is possible to obtain an overall accuracy of 96% using four features from the initial dataset, with sensitivity and specificity values equal to 95% and 99% respectively. Expanding the dataset with interpolated cases, it was possible to improve on these results with 98.8% for accuracy, and 99% for sensitivity and specificity when four features were used.</abstract><type>Journal Article</type><journal>Neurocomputing</journal><volume>228</volume><paginationStart>220</paginationStart><paginationEnd>230</paginationEnd><publisher/><issnPrint>0925-2312</issnPrint><keywords>Physical activity; Overweight; Obesity; Machine learning; Classification; Neural networks; Sensors</keywords><publishedDay>8</publishedDay><publishedMonth>3</publishedMonth><publishedYear>2017</publishedYear><publishedDate>2017-03-08</publishedDate><doi>10.1016/j.neucom.2016.10.040</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>2017-01-17T09:32:57.7310355</lastEdited><Created>2016-11-04T14:44:48.7200237</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>Paul</firstname><surname>Fergus</surname><order>1</order></author><author><firstname>Abir J.</firstname><surname>Hussain</surname><order>2</order></author><author><firstname>John</firstname><surname>Hearty</surname><order>3</order></author><author><firstname>Stuart</firstname><surname>Fairclough</surname><order>4</order></author><author><firstname>Lynne</firstname><surname>Boddy</surname><order>5</order></author><author><firstname>Kelly</firstname><surname>Mackintosh</surname><orcid>0000-0003-0355-6357</orcid><order>6</order></author><author><firstname>Gareth</firstname><surname>Stratton</surname><orcid>0000-0001-5618-0803</orcid><order>7</order></author><author><firstname>Nicky</firstname><surname>Ridgers</surname><order>8</order></author><author><firstname>Dhiya</firstname><surname>Al-Jumeily</surname><order>9</order></author><author><firstname>Ahmed J.</firstname><surname>Aljaaf</surname><order>10</order></author><author><firstname>Jenet</firstname><surname>Lunn</surname><order>11</order></author></authors><documents><document><filename>0030942-04112016144635.pdf</filename><originalFilename>fergus2016.pdf</originalFilename><uploaded>2016-11-04T14:46:35.4570000</uploaded><type>Output</type><contentLength>952426</contentLength><contentType>application/pdf</contentType><version>Accepted Manuscript</version><cronfaStatus>true</cronfaStatus><embargoDate>2017-11-01T00:00:00.0000000</embargoDate><copyrightCorrect>false</copyrightCorrect></document></documents><OutputDurs/></rfc1807>
spelling 2017-01-17T09:32:57.7310355 v2 30942 2016-11-04 A machine learning approach to measure and monitor physical activity in children bdb20e3f31bcccf95c7bc116070c4214 0000-0003-0355-6357 Kelly Mackintosh Kelly Mackintosh true false 6d62b2ed126961bed81a94a2beba8a01 0000-0001-5618-0803 Gareth Stratton Gareth Stratton true false 2016-11-04 STSC The growing trend of obesity and overweight worldwide has reached epidemic proportions with one third of the global population now considered obese. This is having a significant medical impact on children and adults who are at risk of developing osteoarthritis, coronary heart disease and stroke, type 2 diabetes, cancers, respiratory problems, and non-alcoholic fatty liver disease. In an attempt to redress the issue, physical activity is being promoted as a fundamental component for maintaining a healthy lifestyle. Recommendations for physical activity levels are issued by most governments as part of their public health measures. However, current techniques and protocols, including those used in laboratory settings, have been criticised. The main concern is that it is not feasible to use multiple pieces of measurement hardware, such as VO2 masks and heart rate monitors, to monitor children in free-living environments due to weight and encumbrance constraints. This has prompted research in the use of wearable sensing and machine learning technology to produce classifications for specific physical activity events. This paper builds on this approach and presents a supervised machine learning method that utilises data obtained from accelerometer sensors worn by children in free-living environments. Our results show that when using an artificial neural network algorithm it is possible to obtain an overall accuracy of 96% using four features from the initial dataset, with sensitivity and specificity values equal to 95% and 99% respectively. Expanding the dataset with interpolated cases, it was possible to improve on these results with 98.8% for accuracy, and 99% for sensitivity and specificity when four features were used. Journal Article Neurocomputing 228 220 230 0925-2312 Physical activity; Overweight; Obesity; Machine learning; Classification; Neural networks; Sensors 8 3 2017 2017-03-08 10.1016/j.neucom.2016.10.040 COLLEGE NANME Sport and Exercise Sciences COLLEGE CODE STSC Swansea University 2017-01-17T09:32:57.7310355 2016-11-04T14:44:48.7200237 Faculty of Science and Engineering School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Sport and Exercise Sciences Paul Fergus 1 Abir J. Hussain 2 John Hearty 3 Stuart Fairclough 4 Lynne Boddy 5 Kelly Mackintosh 0000-0003-0355-6357 6 Gareth Stratton 0000-0001-5618-0803 7 Nicky Ridgers 8 Dhiya Al-Jumeily 9 Ahmed J. Aljaaf 10 Jenet Lunn 11 0030942-04112016144635.pdf fergus2016.pdf 2016-11-04T14:46:35.4570000 Output 952426 application/pdf Accepted Manuscript true 2017-11-01T00:00:00.0000000 false
title A machine learning approach to measure and monitor physical activity in children
spellingShingle A machine learning approach to measure and monitor physical activity in children
Kelly Mackintosh
Gareth Stratton
title_short A machine learning approach to measure and monitor physical activity in children
title_full A machine learning approach to measure and monitor physical activity in children
title_fullStr A machine learning approach to measure and monitor physical activity in children
title_full_unstemmed A machine learning approach to measure and monitor physical activity in children
title_sort A machine learning approach to measure and monitor physical activity in children
author_id_str_mv bdb20e3f31bcccf95c7bc116070c4214
6d62b2ed126961bed81a94a2beba8a01
author_id_fullname_str_mv bdb20e3f31bcccf95c7bc116070c4214_***_Kelly Mackintosh
6d62b2ed126961bed81a94a2beba8a01_***_Gareth Stratton
author Kelly Mackintosh
Gareth Stratton
author2 Paul Fergus
Abir J. Hussain
John Hearty
Stuart Fairclough
Lynne Boddy
Kelly Mackintosh
Gareth Stratton
Nicky Ridgers
Dhiya Al-Jumeily
Ahmed J. Aljaaf
Jenet Lunn
format Journal article
container_title Neurocomputing
container_volume 228
container_start_page 220
publishDate 2017
institution Swansea University
issn 0925-2312
doi_str_mv 10.1016/j.neucom.2016.10.040
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 The growing trend of obesity and overweight worldwide has reached epidemic proportions with one third of the global population now considered obese. This is having a significant medical impact on children and adults who are at risk of developing osteoarthritis, coronary heart disease and stroke, type 2 diabetes, cancers, respiratory problems, and non-alcoholic fatty liver disease. In an attempt to redress the issue, physical activity is being promoted as a fundamental component for maintaining a healthy lifestyle. Recommendations for physical activity levels are issued by most governments as part of their public health measures. However, current techniques and protocols, including those used in laboratory settings, have been criticised. The main concern is that it is not feasible to use multiple pieces of measurement hardware, such as VO2 masks and heart rate monitors, to monitor children in free-living environments due to weight and encumbrance constraints. This has prompted research in the use of wearable sensing and machine learning technology to produce classifications for specific physical activity events. This paper builds on this approach and presents a supervised machine learning method that utilises data obtained from accelerometer sensors worn by children in free-living environments. Our results show that when using an artificial neural network algorithm it is possible to obtain an overall accuracy of 96% using four features from the initial dataset, with sensitivity and specificity values equal to 95% and 99% respectively. Expanding the dataset with interpolated cases, it was possible to improve on these results with 98.8% for accuracy, and 99% for sensitivity and specificity when four features were used.
published_date 2017-03-08T03:37:43Z
_version_ 1763751665211539456
score 11.016235

Similar Items