Comparison between Mother, ActiGraph wGT3X-BT, and a hand tally for measuring steps at various walking speeds under controlled conditions

Riel, Henrik; Rathleff, Camilla Rams; Kalstrup, Pernille Møller; Madsen, Niels Kragh; Pedersen, Elena Selmar; Pape-Haugaard, Louise; Villumsen, Sidsel

doi:10.7717/peerj.2799

Cited by 15 publications

(30 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Given our results, the iPhone SE, including the apps Health, Accupedo, Pacer, and Runtastic Pedometer, and the Garmin Vivofit 2 are suitable tools for step counting in healthy young as well as middle‐aged individuals. Because these devices demonstrated a higher validity compared to the widely used ActiGraph wGTX+ especially at slow walking speeds, thereby confirming previous findings, they may be a promising alternative for step counting, for example, in patients with chronic diseases and severely reduced ambulatory abilities to monitor adherence to PA‐related therapy recommendations such as step goals. In this context, the phone may be placed in a backpack or shoulder bag for accurate step counting and does not necessarily have to be carried in the pants pocket, thus substantially improving the versatility for objective PA assessment in large‐scale clinical as well as research settings.…”

Section: Discussionsupporting

confidence: 78%

“…The ActiGraph worn on the hip was only accurate for treadmill walking speeds of 4.8 and 6.0 km/h. The detected magnitude of the deviation of 23.8% from the gold standard at 3.2 km/h is comparable to the 26.7% deviation Riel et al 24 found in healthy college students (mean age 27.9 years) and thus clearly questions the suitability of the ActiGraph to be used for step detection in slow walking speeds. In addition, the confirmed inaccuracy of the hip-worn ActiGraph seriously contests the use of this device as the reference method for step counting during activities of daily life and thus puts even most recent research findings 37 into question.…”

Section: Role Of Device Position and Clinical Relevance Of Deviatiosupporting

confidence: 61%

“…36 The iPhones achieved this 1% threshold for 4.8 km/h (shoulder bag and backpack) and 6.0 km/h (all positions) and showed slightly higher, yet still very small deviations from the gold standard (< 3%) for 1.6 km/h (when carried in pants pocket), 3.2 km/h (all positions), and free walking (all positions). According to previous validation studies, [24][25][26][27] a deviation of 3% or less from a hand tally can be classified as clinically irrelevant because the difference between the measured steps and the true steps is then small enough so that the device's error deviation does not have a relevant impact on the classification of the individual's PA behavior. The previously presumed possible impact of the phone's position on the accuracy of step detection was not confirmed F I G U R E 2 Median (interquartile range) number of steps as measured by the different devices and by the gold standard (video recording) at each walking speed on the treadmill as well as during free walking.…”

Section: Role Of Device Position and Clinical Relevance Of Deviatiomentioning

confidence: 99%

See 2 more Smart Citations

Validity of activity trackers, smartphones, and phone applications to measure steps in various walking conditions

Höchsmann

Knaier

Eymann

et al. 2018

Scandinavian Med Sci Sports

110

View full text Add to dashboard Cite

To examine the validity of popular smartphone accelerometer applications and a consumer activity wristband compared to a widely used research accelerometer while assessing the impact of the phone's position on the accuracy of step detection. Twenty volunteers from 2 different age groups (Group A: 18-25 years, n = 10; Group B 45-70 years, n = 10) were equipped with 3 iPhone SE smartphones (placed in pants pocket, shoulder bag, and backpack), 1 Samsung Galaxy S6 Edge (pants pocket), 1 Garmin Vivofit 2 wristband, and 2 ActiGraph wGTX+ devices (worn at wrist and hip) while walking on a treadmill (1.6, 3.2, 4.8, and 6.0 km/h) and completing a walking course. All smartphones included 6 accelerometer applications. Video observation was used as gold standard. Validity was evaluated by comparing each device with the gold standard using mean absolute percentage errors (MAPE). The MAPE of the iPhone SE (all positions) and the Garmin Vivofit was small (<3) for treadmill walking ≥3.2 km/h and for free walking. The Samsung Galaxy and hip-worn ActiGraph showed small MAPE only for treadmill walking at 4.8 and 6.0 km/h and for free walking. The wrist-worn ActiGraph showed high MAPE (17-47) for all walking conditions. The iPhone SE and the Garmin Vivofit 2 are accurate tools for step counting in different age groups and during various walking conditions, even during slow walking. The phone's position does not impact the accuracy of step detection, which substantially improves the versatility for physical activity assessment in clinical and research settings.

show abstract

Section: Discussionsupporting

confidence: 78%

Section: Role Of Device Position and Clinical Relevance Of Deviatiosupporting

confidence: 61%

Section: Role Of Device Position and Clinical Relevance Of Deviatiomentioning

confidence: 99%

See 1 more Smart Citation

Validity of activity trackers, smartphones, and phone applications to measure steps in various walking conditions

Höchsmann

Knaier

Eymann

et al. 2018

Scandinavian Med Sci Sports

110

View full text Add to dashboard Cite

show abstract

“…A recent technical assessment of activity trackers’ accuracy has documented that affordable consumer trackers can compete with advanced medical products when it comes to accuracy. 14 …”

Section: Introductionmentioning

confidence: 99%

Contradictions in digital health engagement: An activity tracker’s ambiguous influence on vulnerable young adults’ engagement in own health

2018

View full text Add to dashboard Cite

ObjectiveActivity trackers are designed to support individuals in monitoring and increasing their physical activity. The use of activity trackers among individuals diagnosed with depression and anxiety has not yet been examined. This pilot study investigates how this target group engages with an activity tracker during a 10-week health intervention aimed to increase their physical activity level and improve their physical and mental health.MethodsTwo groups of 11 young adults (aged 18–29 years) diagnosed with depression or anxiety participated in the digital health intervention. The study used mixed methods to investigate the research question. Quantitative health data were used to assess the intervention’s influence on the participants’ health and qualitative data provided insights into the participants’ digital health experience.ResultsThe study demonstrated an ambiguous influence from the use of an activity tracker with positive physical and mental health results, but a fading and even negative digital health engagement and counterproductive competition.ConclusionsThe ambiguous results identify a need for (1) developing strategies for health professionals to provide supervised use of activity trackers and support the target groups’ abilities to convert health information about physical activity into positive health strategies, and (2) designing alternatives for health promoting IT targeted users who face challenges and need motivation beyond self-tracking and competition.

show abstract

“…The protocol may have caused these low accuracies: movement speeds were self-selected when subjects conducted the above four activities. If these activities had instead been performed on a treadmill or paced at a preset speed ( Harrell et al, 2005 ; Riel et al, 2016 ), classifications may have been more distinct; however, treadmill speeds do not represent the innate characteristics or real physical activities performed by children in daily life. It was observed that some participants walked at equal speeds during WS and WF activities, whereas others ran too fast during the RS protocol.…”

Section: Discussionmentioning

confidence: 99%

Deep learning-based classification with improved time resolution for physical activities of children

Jang

Kim

et al. 2018

PeerJ

View full text Add to dashboard Cite

BackgroundThe proportion of overweight and obese people has increased tremendously in a short period, culminating in a worldwide trend of obesity that is reaching epidemic proportions. Overweight and obesity are serious issues, especially with regard to children. This is because obese children have twice the risk of becoming obese as adults, as compared to non-obese children. Nowadays, many methods for maintaining a caloric balance exist; however, these methods are not applicable to children. In this study, a new approach for helping children monitor their activities using a convolutional neural network (CNN) is proposed, which is applicable for real-time scenarios requiring high accuracy.MethodsA total of 136 participants (86 boys and 50 girls), aged between 8.5 years and 12.5 years (mean 10.5, standard deviation 1.1), took part in this study. The participants performed various movement while wearing custom-made three-axis accelerometer modules around their waists. The data acquired by the accelerometer module was preprocessed by dividing them into small sets (128 sample points for 2.8 s). Approximately 183,600 data samples were used by the developed CNN for learning to classify ten physical activities : slow walking, fast walking, slow running, fast running, walking up the stairs, walking down the stairs, jumping rope, standing up, sitting down, and remaining still.ResultsThe developed CNN classified the ten activities with an overall accuracy of 81.2%. When similar activities were merged, leading to seven merged activities, the CNN classified activities with an overall accuracy of 91.1%. Activity merging also improved performance indicators, for the maximum case of 66.4% in recall, 48.5% in precision, and 57.4% in f1 score . The developed CNN classifier was compared to conventional machine learning algorithms such as the support vector machine, decision tree, and k-nearest neighbor algorithms, and the proposed CNN classifier performed the best: CNN (81.2%) > SVM (64.8%) > DT (63.9%) > kNN (55.4%) (for ten activities); CNN (91.1%) > SVM (74.4%) > DT (73.2%) > kNN (65.3%) (for the merged seven activities).DiscussionThe developed algorithm distinguished physical activities with improved time resolution using short-time acceleration signals from the physical activities performed by children. This study involved algorithm development, participant recruitment, IRB approval, custom-design of a data acquisition module, and data collection. The self-selected moving speeds for walking and running (slow and fast) and the structure of staircase degraded the performance of the algorithm. However, after similar activities were merged, the effects caused by the self-selection of speed were reduced. The experimental results show that the proposed algorithm performed better than conventional algorithms. Owing to its simplicity, the proposed algorithm could be applied to real-time applicaitons.

show abstract

Comparison between Mother, ActiGraph wGT3X-BT, and a hand tally for measuring steps at various walking speeds under controlled conditions

Cited by 15 publications

References 36 publications

Validity of activity trackers, smartphones, and phone applications to measure steps in various walking conditions

Validity of activity trackers, smartphones, and phone applications to measure steps in various walking conditions

Contradictions in digital health engagement: An activity tracker’s ambiguous influence on vulnerable young adults’ engagement in own health

Deep learning-based classification with improved time resolution for physical activities of children

Contact Info

Product

Resources

About