Evaluation of Neural Networks to Identify Types of Activity Using Accelerometers

Vries, S.I. de; Garre, Francisca Galindo; Engbers, L.H.; Hildebrandt, V.H.; Buuren, Stef van

doi:10.1249/mss.0b013e3181e5797d

Cited by 67 publications

(95 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In a study carried out by De Vries et al a series of ANNs were developed to predict PA in children across a range of activity types. However, the results reported were significantly lower than those reported in Staudenmayer et al [37] with classification accuracies between 57.2% and 76.8% [30], [31].…”

Section: Physical Activity Classificationcontrasting

confidence: 66%

“…Data from 100 children was used, comprising of data for 12 activities each of child performed. The best performing design was trained using features extracted from a range of time windows (10,15,20,30 and 60 seconds). This rigorous methodology yielded particularly good results, with the most successful network able to predict PA type with 88.4% accuracy over a 60-second window.…”

Section: Physical Activity Classificationmentioning

confidence: 99%

See 1 more Smart Citation

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

Fergus

Hussain

Hearty³

et al. 2017

Neurocomputing

View full text Add to dashboard Cite

ABSTRACT-Physical Activity is a fundamental component for the maintenance of a healthy lifestyle. Recommendations for physical activity levels are issued by most governments as part of public health measures. Therefore, it is vital for regulatory purposes, that there are reliable measurements of physical activity. However, the techniques and protocols used in existing physical activity research, including laboratory-based measurement, have received increasingly critical scrutiny in recent times. Consequently, physical activity researchers have begun to explore the use of wearable sensing technology to capture large amounts of data and the use of machine learning techniques, specifically artificial neural networks, to produce classifications for specific physical activity events. This paper explores this idea further and presents a supervised machine learning approach that utilises data obtained from accelerometer sensors worn by children in free-living environments.The paper posits a rigorous data science approach that presents a set of activities and features suitable for measuring physical activity in children in free-living environments. A Multilayer Perceptron neural network is used to classify physical activities by activity type, using ecologically valid data from body worn accelerometer sensors. A rigorous reproducible data science methodology is presented for subsequent use in physical activity research. Our results show that it was possible to obtain an overall accuracy of 92% using the initial data set, and 99.8% using interpolated cases.

show abstract

Section: Physical Activity Classificationcontrasting

confidence: 66%

Section: Physical Activity Classificationmentioning

confidence: 99%

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

Fergus

Hussain

Hearty³

et al. 2017

Neurocomputing

View full text Add to dashboard Cite

show abstract

“…In a study carried out by De Vries et al, a series of ANNs were developed to predict PA in children across a range of activity types. However, the results reported were significantly lower than those reported in (Staudenmayer et al, 2009) with classification accuracies between 57.2% and 76.8% (De Vries et al, 2011;De Vries, Garre, Engbers, H., & Van Buuren, 2001). …”

Section: Introductioncontrasting

confidence: 64%

A Machine Learning Approach to Measure and Monitor Physical Activity in Children to Help Fight Overweight and Obesity

Fergus

Hussain

Hearty

et al. 2015

Intelligent Computing Theories and Methodologies

View full text Add to dashboard Cite

Abstract. Physical Activity is important for maintaining healthy lifestyles. Recommendations for physical activity levels are issued by most governments as part of public health measures. As such, reliable measurement of physical activity for regulatory purposes is vital. This has lead research to explore standards for achieving this using wearable technology and artificial neural networks that produce classifications for specific physical activity events. Applied from a very early age, the ubiquitous capture of physical activity data using mobile and wearable technology may help us to understand how we can combat childhood obesity and the impact that this has in later life. A supervised machine learning approach is adopted in this paper that utilizes data obtained from accelerometer sensors worn by children in free-living environments. The paper presents a set of activities and features suitable for measuring physical activity and evaluates the use of a Multilayer Perceptron neural network to classify physical activities by activity type. A rigorous reproducible data science methodology is presented for subsequent use in physical activity research. Our results 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 and four feature combinations were used.

show abstract

“…6,8,9,14 Machine learning has experienced an increase in popularity in sporting and exercise research, with applications such as the prediction of competition outcome 15 and quantification of movement activity types. 16,17 This increase in popularity has stemmed from the potential ability of these approaches to account for nonlinearity within datasets and thus display improved performance. 18 Therefore, implementation of such approaches may help to better understand TL at the individual level, thus increasing the accuracy in understanding the relationship between internal and external load, and thus, athlete management.…”

mentioning

confidence: 99%

Relationships Between Internal and External Training Load in Team-Sport Athletes: Evidence for an Individualized Approach

Bartlett¹,

O’Connor²,

Pitchford³

et al. 2017

International Journal of Sports Physiology and Performance

149

119

View full text Add to dashboard Cite

Purpose The aim of this study was to quantify and predict relationships between RPE and GPS training load variables in professional Australian Football (AF) players using group and individualised modelling approaches. Methods Training load data (GPS and RPE) for 41 professional AF players was obtained over a period of 27 weeks. A total of 2711 training observations were analysed with a total of 66 13 sessions per player (range; 39 to 89).Separate generalised estimating equations (GEE) and artificial neural network analyses (ANN) were conducted to determine the ability to predict RPE from training load variables (i.e. session distance, high-speed running (HSR), high-speed running %, m·min Further, importance plots generated from the ANN revealed session distance was most predictive of RPE in 36 of the 41 players, whereas, HSR was predictive of RPE in just 3 players and m . min -1 as predictive as session distance in just 2 players. Conclusions This study demonstrates that machine learning approaches may outperform more traditional methodologies with respect to predicting athlete responses to training load. These approaches enable further individualisation of load monitoring, leading to more accurate training prescription and evaluation.

show abstract

Evaluation of Neural Networks to Identify Types of Activity Using Accelerometers

Abstract: Relatively simple ANN models perform well in identifying the type but not the speed of the activity of adults from accelerometer data.

Cited by 67 publications

References 17 publications

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

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

A Machine Learning Approach to Measure and Monitor Physical Activity in Children to Help Fight Overweight and Obesity

Relationships Between Internal and External Training Load in Team-Sport Athletes: Evidence for an Individualized Approach

Contact Info

Product

Resources

About