Detection of Type, Duration, and Intensity of Physical Activity Using an Accelerometer

Bonomi, A.; Goris, Annelies H. C.; Yin, Bin

doi:10.1249/mss.0b013e3181a24536

Cited by 207 publications

(204 citation statements)

References 31 publications

Supporting

Mentioning

199

Contrasting

Order By: Relevance

“…Differences in performance between the classification models are small, with the SVM and the majority voting scheme achieving slightly higher accuracy. Other studies using similar schemes have reported accuracies that are comparable to what is presented in this study (93% [25], 94-100% [24], 90.1% [26]). Although all models showed high aggregated performance when evaluated using a leaveone-subject-out cross-validation, the class-specific F-scores revealed that one particular class was difficult to distinguish from the others.…”

Section: Discussionsupporting

confidence: 89%

“…Studies have also focused on using more unobtrusive sensors to detect simpler activities. Classification models, when trained and evaluated using data collected with a single accelerometer in laboratory sessions, have shown promising performance and a high success rate: 94-100% [24], 93% [25], 90.8% [26] and 89% [19]. However, the reproducibility of the performance achieved by these models in daily life is currently unknown.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Identifying Types of Physical Activity With a Single Accelerometer: Evaluating Laboratory-trained Algorithms in Daily Life

Gyllensten

Bonomi

2011

IEEE Trans. Biomed. Eng.

158

118

View full text Add to dashboard Cite

Abstract-Accurate identification of physical activity types has been achieved in laboratory conditions using single-site accelerometers and classification algorithms. This methodology is then applied to free-living subjects to determine activity behaviour. This study aimed at analysing the reproducibility of the accuracy of laboratory-trained classification algorithms in free-living subjects during daily life. A support vector machine (SVM), a feed-forward neural network (NN) and a decision tree (DT) were trained with data collected by a waist-mounted accelerometer during a laboratory trial. The reproducibility of the classification performance was tested on data collected in daily life using a multiple-site accelerometer (IDEEA) augmented with an activity diary for 20 healthy subjects (age: 30 ± 9; BMI: 23.0 ± 2.6 kg/m 2 ). Leave-one-subject-out cross-validation of the training data showed accuracies of 95.1 ± 4.3%, 91.4 ± 6.7% and 92.2 ± 6.6% for the SVM, NN and DT, respectively. All algorithms showed a significantly decreased accuracy in daily life as compared to the reference truth represented by the IDEEA and diary classifications (75.6 ± 10.4%, 74.8 ± 9.7%, and 72.2 ± 10.3%; p<0.05). In conclusion, cross-validation of training data overestimates the accuracy of the classification algorithms in daily life.Index Terms-Assessment of daily physical activity, classification algorithms, IDEEA, physical activity, triaxial accelerometer

show abstract

Section: Discussionsupporting

confidence: 89%

Section: Introductionmentioning

confidence: 99%

Identifying Types of Physical Activity With a Single Accelerometer: Evaluating Laboratory-trained Algorithms in Daily Life

Gyllensten

Bonomi

2011

IEEE Trans. Biomed. Eng.

158

118

View full text Add to dashboard Cite

show abstract

“…Advances have, however, been made in accelerometer data processing with the development of more sophisticated approaches to data modelling analysis (Bonomi et al 2009, Pober et al 2006, Staudenmayer et al 2009, Zhang et al 2003. This area warrants further investigation in studies of pre-school children to determine if this will offer an accurate means of classifying physical activity behaviour.…”

Section: Discussionmentioning

confidence: 99%

Validity of wrist worn accelerometers and comparability between hip and wrist placement sites in estimating physical activity behaviour in preschool children

Hislop

Palmer²,

Anand

et al. 2016

Physiol. Meas.

View full text Add to dashboard Cite

Wrist-worn accelerometers can increase compliance with wearing accelerometers, however, several large scale studies continue to use hip-worn accelerometers and it is unclear how comparable data is from the two sites. The study aims were: to investigate agreement between wrist-and hip-worn accelerometers and to determine the validity of Johansson et al cut-points for wrist worn accelerometers in preschool children.A sample of 32 preschool children (21 boys, 4.2 (0.5) years, BMI 16.6 (1.1)) were videoed wearing GT3X+ accelerometers on their wrist and hip while they engaged in 1 hour of free-play in their nursery.Children's activity were coded using, the Children's Activity Rating Scale (CARS): with CARS, level 1 'sedentary' and levels 2 to 5 were classified as time spent in TPA. Accelerometry data were processed using Johansson et al cut-points for the wrist data and Evenson et al cut-points for the hip data, into time spent in different intensities of PA. The mean counts per minute (cpm) from the hip and wrist were compared.There was a strong correlation between the hip and wrist cpm (r=0.81, p<0.01) and total count data (r=0.83 p<0.01), however there was a large systematic bias with wide limits of agreement. Good agreement (mean difference (LOA) 1.1 (-9.9, 12.1) was found between the CARS estimate of TPA (29.

show abstract

“…We showed that identifying three different postures, that is, lying, sitting and standing, and three types of locomotion movements, that is, walking, running and cycling, was accurately achieved using a single accelerometer positioned at the waist. 54 However, standing was often confused with sitting and for this reason in later studies 55 the sitting and standing classes were combined together in a single category describing static events. Multiple accelerometers allow the successful distinction between sitting and standing.…”

Section: Activity Recognition Using Multiple or Single-site Accelerommentioning

confidence: 99%

Advances in physical activity monitoring and lifestyle interventions in obesity: a review

Bonomi

Westerterp

2011

Int J Obes

View full text Add to dashboard Cite

Obesity represents a strong risk factor for developing chronic diseases. Strategies for disease prevention often promote lifestyle changes encouraging participation in physical activity. However, determining what amount of physical activity is necessary for achieving specific health benefits has been hampered by the lack of accurate instruments for monitoring physical activity and the related physiological outcomes. This review aims at presenting recent advances in activity-monitoring technology and their application to support interventions for health promotion. Activity monitors have evolved from step counters and measuring devices of physical activity duration and intensity to more advanced systems providing quantitative and qualitative information on the individuals' activity behavior. Correspondingly, methods to predict activity-related energy expenditure using bodily acceleration and subjects characteristics have advanced from linear regression to innovative algorithms capable of determining physical activity types and the related metabolic costs. These novel techniques can monitor modes of sedentary behavior as well as the engagement in specific activity types that helps to evaluate the effectiveness of lifestyle interventions. In conclusion, advances in activity monitoring have the potential to support the design of response-dependent physical activity recommendations that are needed to generate effective and personalized lifestyle interventions for health promotion.

show abstract

Detection of Type, Duration, and Intensity of Physical Activity Using an Accelerometer

Abstract: This study demonstrated the ability of a triaxial accelerometer in detecting type, duration, and intensity of physical activity using models based on acceleration features. Future studies are needed to validate the presented models in free-living conditions.

Cited by 207 publications

References 31 publications

Identifying Types of Physical Activity With a Single Accelerometer: Evaluating Laboratory-trained Algorithms in Daily Life

Identifying Types of Physical Activity With a Single Accelerometer: Evaluating Laboratory-trained Algorithms in Daily Life

Validity of wrist worn accelerometers and comparability between hip and wrist placement sites in estimating physical activity behaviour in preschool children

Advances in physical activity monitoring and lifestyle interventions in obesity: a review

Contact Info

Product

Resources

About