Conference Paper/Proceeding/Abstract 721 views
A Machine Learning Approach to Measure and Monitor Physical Activity in Children to Help Fight Overweight and Obesity
Intelligent Computing Theories and Methodologies, Volume: 9226, Pages: 676 - 688
Full text not available from this repository: check for access using links below.
Abstract. Physical Activity is important for maintaining healthy lifestyles.Recommendations for physical activity levels are issued by most governmentsas part of public health measures. As such, reliable measurement of physicalactivity for regulatory purposes is vital. This has lead research to expl...
|Published in:||Intelligent Computing Theories and Methodologies|
Check full text
No Tags, Be the first to tag this record!
Abstract. Physical Activity is important for maintaining healthy lifestyles.Recommendations for physical activity levels are issued by most governmentsas part of public health measures. As such, reliable measurement of physicalactivity for regulatory purposes is vital. This has lead research to explorestandards for achieving this using wearable technology and artificial neuralnetworks that produce classifications for specific physical activity events.Applied from a very early age, the ubiquitous capture of physical activity datausing mobile and wearable technology may help us to understand how we cancombat childhood obesity and the impact that this has in later life. A supervisedmachine learning approach is adopted in this paper that utilizes data obtainedfrom accelerometer sensors worn by children in free-living environments. Thepaper presents a set of activities and features suitable for measuring physicalactivity and evaluates the use of a Multilayer Perceptron neural network toclassify physical activities by activity type. A rigorous reproducible data sciencemethodology is presented for subsequent use in physical activity research. Ourresults show that it was possible to obtain an overall accuracy of 96 % with 95 %for sensitivity, 99 % for specificity and a kappa value of 94 % when three andfour feature combinations were used.
Physical activity, Overweight, Obesity, Machine learning, Neural networks, Sensors
College of Engineering