scholarly journals Criterion-Validity of Commercially Available Physical Activity Tracker to Estimate Step Count, Covered Distance and Energy Expenditure during Sports Conditions

2017 ◽  
Vol 8 ◽  
Author(s):  
Yvonne Wahl ◽  
Peter Düking ◽  
Anna Droszez ◽  
Patrick Wahl ◽  
Joachim Mester
Keyword(s):  
Physical Activity ◽  
Energy Expenditure ◽  
Criterion Validity ◽  
Step Count ◽  
Activity Tracker

scholarly journals Neural Network–Based Algorithm for Adjusting Activity Targets to Sustain Exercise Engagement Among People Using Activity Trackers: Retrospective Observation and Algorithm Development Study

10.2196/18142 ◽  
2020 ◽  
Vol 8 (9) ◽  
pp. e18142
Author(s):  
Ramin Mohammadi ◽  
Mursal Atif ◽  
Amanda Jayne Centi ◽  
Stephen Agboola ◽  
Kamal Jethwani ◽  
...  
Keyword(s):  
Neural Network ◽  
Physical Activity ◽  
Machine Learning ◽  
Learning Model ◽  
Step Count ◽  
Daily Step ◽  
Activity Tracker ◽  
Activity Trackers ◽  
Machine Learning Model ◽  
Daily Step Count

Background It is well established that lack of physical activity is detrimental to the overall health of an individual. Modern-day activity trackers enable individuals to monitor their daily activities to meet and maintain targets. This is expected to promote activity encouraging behavior, but the benefits of activity trackers attenuate over time due to waning adherence. One of the key approaches to improving adherence to goals is to motivate individuals to improve on their historic performance metrics. Objective The aim of this work was to build a machine learning model to predict an achievable weekly activity target by considering (1) patterns in the user’s activity tracker data in the previous week and (2) behavior and environment characteristics. By setting realistic goals, ones that are neither too easy nor too difficult to achieve, activity tracker users can be encouraged to continue to meet these goals, and at the same time, to find utility in their activity tracker. Methods We built a neural network model that prescribes a weekly activity target for an individual that can be realistically achieved. The inputs to the model were user-specific personal, social, and environmental factors, daily step count from the previous 7 days, and an entropy measure that characterized the pattern of daily step count. Data for training and evaluating the machine learning model were collected over a duration of 9 weeks. Results Of 30 individuals who were enrolled, data from 20 participants were used. The model predicted target daily count with a mean absolute error of 1545 (95% CI 1383-1706) steps for an 8-week period. Conclusions Artificial intelligence applied to physical activity data combined with behavioral data can be used to set personalized goals in accordance with the individual’s level of activity and thereby improve adherence to a fitness tracker; this could be used to increase engagement with activity trackers. A follow-up prospective study is ongoing to determine the performance of the engagement algorithm.

scholarly journals Criterion validity of a wrist-worn activity tracker in laboratory and free-living setting in patients with chronic pain (Preprint)

2020 ◽  
Author(s):  
Veronica Sjöberg ◽  
Jens Westergren ◽  
Andreas Monnier ◽  
Ricardo LoMartire ◽  
Maria Hagströmer ◽  
...  
Keyword(s):  
Chronic Pain ◽  
Criterion Validity ◽  
Step Count ◽  
Fair Agreement ◽  
Laboratory Setting ◽  
Poor Agreement ◽  
Activity Tracking ◽  
Free Living ◽  
Activity Tracker ◽  
Tracking Devices

BACKGROUND Physical Activity (PA) is evidently a crucial part of the rehabilitation process for patients suffering from chronic pain. Modern wrist-worn activity tracking devices seemingly have a great potential to provide objective feedback and assist in the adoption of healthy PA behavior by supplying data of energy expenditure expressed as Metabolic Equivalents (METS). However, no studies have been found of any wrist-worn activity tracking devices’ criterion validity in estimating METS, heart rate (HR), or step count in patients with chronic pain. OBJECTIVE The aim was to determine the criterion validity of wrist-worn activity tracking devices for estimations of METS, HR, and step count in a controlled laboratory setting and free-living settings for patients with chronic pain. METHODS In this combined laboratory and field validation study, METS, HR, and step count were simultaneously estimated by a wrist-worn activity tracker (Fitbit Versa), indirect calorimetry (Jaeger Oxycon Pro), and a research-grade hip-worn accelerometer (ActiGraph GT3X) during a treadmill walk at three speeds (3.0, 4.5, and 6.0 km/h) in a laboratory setting. METS and step count were also estimated by the wrist-worn activity tracker in free-living settings for 72 hours. The criterion validity was determined by conventional statistics (ICC and Spearman rho) and graphical plots (Bland-Altman Plots) as well as by Mean Absolute Percentage Error (MAPE). Analysis of Variance (ANOVA) was used to determine any significant systematic differences between estimations. RESULTS A total of 42 patients (76% females), 25-66 years of age, with chronic pain, were included. Results showed that the wrist-worn activity tracking devices (Fitbit Versa) systematically overestimated METS when compared to the criterion measurement (Jaeger Oxycon Pro) and the relative criterion measurement (ActiGraph GT3X). Poor agreement and correlation was shown in estimated METS between Fitbit Versa and both Jaeger Oxycon Pro and ActiGraph GT3X at all treadmill speeds. Estimations of HR emerged with poor to fair agreement during laboratory-based treadmill walks. For step count, the wrist-worn devices showed a fair agreement and fair correlation at most treadmill speeds. In free-living settings, however, the agreement of step count between wrist-worn devices and waist-worn accelerometer was good, and the correlation was excellent. CONCLUSIONS The wrist-worn device systematically overestimated METS and showed poor agreement and correlation compared to the criterion measurement (Jaeger Oxycon Pro) and the relative criterion measurement (ActiGraph GT3X), which needs to be considered when used clinically. Step count measured from the wrist, however, seemed to be a valid estimation, suggesting that future guidelines could include such variables in this group with chronic pain. CLINICALTRIAL Not applicable in this study

Validation of a Noninvasive, Disposable Activity Monitor for Clinical Applications

2017 ◽  
Vol 14 (7) ◽  
pp. 546-551 ◽  
Author(s):  
Greg Welk ◽  
Youngwon Kim ◽  
Robin P. Shook ◽  
Laura Ellingson ◽  
Roberto L. Lobelo
Keyword(s):  
Physical Activity ◽  
Energy Expenditure ◽  
Physical Activity Behavior ◽  
Activity Monitor ◽  
Criterion Validity ◽  
Sensewear Armband ◽  
Statistical Equivalence ◽  
Activity Behavior ◽  
Highly Correlated ◽  
And Behavior

Background:The study evaluated the concurrent and criterion validity of a new, disposable activity monitor designed to provide objective data on physical activity and energy expenditure in clinical populations.Methods:A sample of healthy adults (n = 52) wore the disposable Metria IH1 along with the established Sensewear armband (SWA) monitor for a 1-week period. Concurrent validity was examined by evaluating the statistical equivalence of estimates from the Metria and the SWA. Criterion validity was examined by comparing the relative accuracy of the Metria IH1 and the SWA for assessing walking/running. The absolute validity of the 2 monitors was compared by computing correlations and mean absolute percent error (MAPE) relative to criterion data from a portable metabolic analyzer.Results:The output from 2 monitors was highly correlated (correlations > 0.90) and the summary measures yielded nearly identical allocations of time spent in physical activity and energy expenditure. The monitors yielded statistically equivalent estimates and had similar absolute validity relative to the criterion measure (12% to 15% error).Conclusions:The disposable nature of the adhesive Metria IH1 monitor offers promise for clinical evaluation of physical activity behavior in patients. Additional research is needed to test utility for counseling and behavior applications.

scholarly journals Evaluation of the FitBark Activity Monitor for Measuring Physical Activity in Dogs

Animals ◽  
2021 ◽  
Vol 11 (3) ◽  
pp. 781
Author(s):  
Jessica Colpoys ◽  
Dean DeCock
Keyword(s):  
Physical Activity ◽  
Video Analysis ◽  
Activity Monitor ◽  
Criterion Validity ◽  
Step Count ◽  
Phase 3 ◽  
Activity Data ◽  
Phase Approach ◽  
Low Correlation ◽  
Three Phase

Accelerometers track changes in physical activity which can indicate health and welfare concerns in dogs. The FitBark 2 (FitBark) is an accelerometer for use with dogs; however, no studies have externally validated this tool. The objective of this study was to evaluate FitBark criterion validity by correlating FitBark activity data to dog step count. Dogs (n = 26) were fitted with a collar-mounted FitBark and individually recorded for 30 min using a three-phase approach: (1) off-leash room explore; (2) human–dog interaction; and (3) on-leash walk. Video analysis was used to count the number of times the front right paw touched the ground (step count). Dog step count and FitBark activity were moderately correlated across all phases (r = 0.65, p < 0.001). High correlations between step count and FitBark activity were observed during phases 1 (r = 0.795, p < 0.001) and 2 (r = 0.758, p < 0.001), and a low correlation was observed during phase 3 (r = 0.498, p < 0.001). In conclusion, the FitBark is a valid tool for tracking physical activity in off-leash dogs; however, more work should be done to identify the best method of tracking on-leash activity.

Text messaging and activity tracker motivation program to increase physical activity in cancer survivors.

2018 ◽  
Vol 36 (7_suppl) ◽  
pp. 92-92
Author(s):  
Bridget F. Koontz ◽  
Erica Levine ◽  
Frances McSherry ◽  
Tykeytra Dale ◽  
Martin Streicher ◽  
...  
Keyword(s):  
Physical Activity ◽  
Cancer Survivors ◽  
Text Messaging ◽  
Text Message ◽  
Application Programming Interface ◽  
Feedback System ◽  
Activity Level ◽  
Step Count ◽  
Activity Tracker ◽  
Application Programming

92 Background: Cancer survivors have high rates of sedentary behavior leading to obesity and cardiovascular disease. Physical activity improves quality of life (QOL) and reduces morbidity and mortality. However, cancer survivors commonly cite motivation as a barrier to increasing physical activity. We hypothesized that a motivational text-messaging feedback system linked to a Fitbit Flex activity tracker would increase the activity level of survivors and those undergoing cancer treatment. Methods: 29 participants were enrolled in an IRB-approved single-institution study. Eligibility allowed any cancer/stage, ≤2 days of exercise per week, life expectancy of 12+ months, and smartphone access. After baseline fitness/QOL testing, participants were provided a Fitbit Flex activity tracker. A text-messaging program automatically uploaded data from the tracker via an application programming interface and provided personalized text message feedback to subject’s smartphone daily for 3 months. Primary endpoint was change in step count from baseline to 3 months, with additional endpoints of change in 6 minute walk/QOL measures at 3 months, and continued exercise/use of tracker at 6 months. Results: To date, 24 have completed the 3 month program. Both academic and community sites participated, including areas with limited internet access. Most participants were female (71%) and white (63%). Eight cancer types and all stages were represented. Three participants withdrew – one because of lost tracker, one cancer death, and one “disappointed” with tracker function. Median daily steps at baseline were 3773 (IQR 2928) and 4365 at 3 months (IQR 4864). 42% had at least a 20% increase in median step count at 3 months. Improvement was noted in 45% of survivors and 38% of active treatment participants. Participants frequently used research nurses for guidance on use of wearable tracker (e.g. syncing, charging, features). Conclusions: Activity tracker with personalized daily feedback via text message successfully motivates cancer patients to increase daily activity. Patients are interested in health technology, but require technical support and coaching to maintain use. Clinical trial information: NCT02627079.

scholarly journals Study of Factors Affecting the Step Count and Physical Activity Related Energy Expenditure of Healthy Elderly: Is It Possible to Assess Physical Activity by Only The Step Count?

2015 ◽  
Vol 30 (4) ◽  
pp. 513-518
Author(s):  
Kazuaki SUYAMA ◽  
Masaharu ASAI ◽  
Takako TANAKA ◽  
Kenichirou TANAKA ◽  
Naomi MIYAMOTO ◽  
...  
Keyword(s):  
Physical Activity ◽  
Energy Expenditure ◽  
Step Count ◽  
Factors Affecting ◽  
Healthy Elderly ◽  
Related Energy

scholarly journals Examining physical activity and quality of life in adults with autism spectrum disorder and intellectual disability

2021 ◽  
pp. 174462952110334
Author(s):  
Brianne Tomaszewski ◽  
Melissa N Savage ◽  
Kara Hume
Keyword(s):  
Quality Of Life ◽  
Physical Activity ◽  
Intellectual Disability ◽  
Step Count ◽  
Daily Step ◽  
Adults With Autism ◽  
Activity Tracker ◽  
Daily Step Count ◽  
The Relationship

Adults with autism and co-occurring intellectual disability engage in low levels of physical activity and are at increased risk of developing secondary health conditions attributed to physical inactivity compared to adults in the general population. Few studies have examined the use of objective measures to characterize physical activity levels for adults with autism and intellectual disability. The current study aimed to examine the relationship between physical activity, using an activity tracker, and quality of life in adults with autism and intellectual disability. In the current study, 38 adults with autism and intellectual disability, ages 18–55, wore a Fitbit Flex 2® activity tracker for 1 week, and completed the Quality of Life Questionnaire. The relationship between average daily step count quality of life was examined. Most adults in the sample were overweight and taking fewer daily steps than recommended guidelines. Increased average daily step count was significantly associated with quality of life.

scholarly journals Neural Network–Based Algorithm for Adjusting Activity Targets to Sustain Exercise Engagement Among People Using Activity Trackers: Retrospective Observation and Algorithm Development Study (Preprint)

2020 ◽  
Author(s):  
Ramin Mohammadi ◽  
Mursal Atif ◽  
Amanda Jayne Centi ◽  
Stephen Agboola ◽  
Kamal Jethwani ◽  
...  
Keyword(s):  
Neural Network ◽  
Physical Activity ◽  
Machine Learning ◽  
Learning Model ◽  
Step Count ◽  
Daily Step ◽  
Activity Tracker ◽  
Activity Trackers ◽  
Machine Learning Model ◽  
Daily Step Count

BACKGROUND It is well established that lack of physical activity is detrimental to the overall health of an individual. Modern-day activity trackers enable individuals to monitor their daily activities to meet and maintain targets. This is expected to promote activity encouraging behavior, but the benefits of activity trackers attenuate over time due to waning adherence. One of the key approaches to improving adherence to goals is to motivate individuals to improve on their historic performance metrics. OBJECTIVE The aim of this work was to build a machine learning model to predict an achievable weekly activity target by considering (1) patterns in the user’s activity tracker data in the previous week and (2) behavior and environment characteristics. By setting realistic goals, ones that are neither too easy nor too difficult to achieve, activity tracker users can be encouraged to continue to meet these goals, and at the same time, to find utility in their activity tracker. METHODS We built a neural network model that prescribes a weekly activity target for an individual that can be realistically achieved. The inputs to the model were user-specific personal, social, and environmental factors, daily step count from the previous 7 days, and an entropy measure that characterized the pattern of daily step count. Data for training and evaluating the machine learning model were collected over a duration of 9 weeks. RESULTS Of 30 individuals who were enrolled, data from 20 participants were used. The model predicted target daily count with a mean absolute error of 1545 (95% CI 1383-1706) steps for an 8-week period. CONCLUSIONS Artificial intelligence applied to physical activity data combined with behavioral data can be used to set personalized goals in accordance with the individual’s level of activity and thereby improve adherence to a fitness tracker; this could be used to increase engagement with activity trackers. A follow-up prospective study is ongoing to determine the performance of the engagement algorithm.

scholarly journals The validity of activity trackers is affected by walking speed: the criterion validity of Garmin Vivosmart® HR and StepWatch™ 3 for measuring steps at various walking speeds under controlled conditions

PeerJ ◽  
2020 ◽  
Vol 8 ◽  
pp. e9381
Author(s):  
Frederik Rose Svarre ◽  
Mads Møller Jensen ◽  
Josephine Nielsen ◽  
Morten Villumsen
Keyword(s):  
Criterion Validity ◽  
Step Count ◽  
Healthcare Personnel ◽  
Percentage Error ◽  
Activity Tracker ◽  
Activity Trackers ◽  
Controlled Conditions ◽  
Healthcare It ◽  
Graded Activity ◽  
Walking Speeds

Introduction The use of activity trackers has increased both among private consumers and in healthcare. It is therefore relevant to consider whether a consumer-graded activity tracker is comparable to or may substitute a research-graded activity tracker, which could further increase the use of activity trackers in healthcare and rehabilitation. Such use will require knowledge of their accuracy as the clinical implications may be significant. Studies have indicated that activity trackers are not sufficiently accurate, especially at lower walking speeds. The present study seeks to inform decision makers and healthcare personnel considering implementing physical activity trackers in clinical practice. This study investigates the criterion validity of the consumer-graded Garmin Vivosmart® HR and the research-graded StepWatch™ 3 compared with manual step count (gold standard) at different walking speeds under controlled conditions. Methods Thirty participants, wearing Garmin Vivosmart® HR at the wrist and StepWatch™ 3 at the ankle, completed six trials on a treadmill at different walking speeds: 1.6 km/h, 2.4 km/h, 3.2 km/h, 4.0 km/h, 4.8 km/h, and 5.6 km/h. The participants were video recorded, and steps were registered by manual step count. Medians and inter-quartile ranges (IQR) were calculated for steps and differences in steps between manually counted steps and the two devices. In order to assess the clinical relevance of the tested devices, the mean absolute percentage error (MAPE) was determined at each speed. A MAPE ≤3% was considered to be clinically irrelevant. Furthermore, differences between manually counted steps and steps recorded by the two devices were presented in Bland–Altman style plots. Results The median of differences in steps between Garmin Vivosmart® HR and manual step count ranged from −49.5 (IQR = 101) at 1.6 km/h to −1 (IQR = 4) at 4.0 km/h. The median of differences in steps between StepWatch™ 3 and manual step count were 4 (IQR = 14) at 1.6 km/h and 0 (IQR = 1) at all other walking speeds. The results of the MAPE showed that differences in steps counted by Garmin Vivosmart® HR were clinically irrelevant at walking speeds 3.2–4.8 km/h (MAPE: 0.61–1.27%) as the values were below 3%. Differences in steps counted by StepWatch™ 3 were clinically irrelevant at walking speeds 2.4–5.6 km/h (MAPE: 0.08–0.35%). Conclusion Garmin Vivosmart® HR tended to undercount steps compared with the manual step count, and StepWatch™ 3 slightly overcounted steps compared with the manual step count. Both the consumer-graded activity tracker (Garmin Vivosmart® HR) and the research-graded (StepWatch™ 3) are valid in detecting steps at selected walking speeds in healthy adults under controlled conditions. However, both activity trackers miscount steps at slow walking speeds, and the consumer graded activity tracker also miscounts steps at fast walking speeds.

Sign in / Sign up

Export Citation Format

Share Document