Advantages and Limitations of Wearable Activity Trackers: Considerations for Patients and Clinicians

Walker, Rachel; Hickey, Amanda; Freedson, Patty S.

doi:10.1188/16.cjon.606-610

Cited by 41 publications

(35 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A major limitation of recent studies on the validity testing of actigraphs was that most evaluations of the intensity of daily PAs have been conducted in laboratories, whereas actigraphs are usually applied in free-living environments [8,22]. Because of the favorable ecological validity of field experiments [45], in the present study, a randomized controlled trial was conducted that assigned participants with different experimental treatments in actual environments.…”

Section: Methodsologymentioning

confidence: 99%

“…Recent studies have indicated that although the validity for step counting of consumer-grade activity trackers is acceptable [32], they may have lower validity for energy expenditure (EE) and sleep compared with other criterion measures [22]. Low accuracy in activity tracking might mislead and jeopardize professional clinical judgments, because these trackers are widely used as a reference for improving public health [8]. Research-based actigraphs are conventionally and widely used in medical evaluations of sleep and rest, such as the studies of circadian rest-activity cycles by Brown et al [33] and wake-sleep conditions by Cole et al [34] and Matsumoto et al [35].…”

Section: Literature Reviewmentioning

confidence: 99%

“…Although portable activity trackers may be useful for monitoring users’ physical behavior, the accuracy of these devices is arguable, and they could cause errors in the evaluation of daily exercise or in clinical judgment [8]. Presently, many people, including professional athletes, use wearable devices, mostly consumer-grade activity trackers, to record and appraise their PA.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Validity of Research-Grade Actigraphy Unit for Measuring Exercise Intensity

Han

Wang

2017

IJERPH

View full text Add to dashboard Cite

This study was conducted in a free-living setting to investigate the measurement validity of a research-based actigraph for strolling and jogging, and to provide a reference for actual practice and research. Because inadequate physical activity (PA) or sedentary lifestyle has become the fourth leading risk factor for mortality worldwide, many countries have been vigorously promoting the concept of “active living”, and the public has been investing greater effort into intensifying their PA. Although research-grade actigraphs have been widely applied to evaluate PA in routine environments, the measurement results may not accurately reflect the wearers’ PA. Unlike most relevant research, which is conducted in well-controlled laboratory environments, the present study was implemented in the field to examine the sensitivity and convergent validity of the MicroMini Motionlogger® Actigraph during strolling and jogging. The following results were revealed: (1) Although the exercise movement speed while jogging was significantly faster than that while strolling, the actigraph readings showed no significant difference between strolling and jogging; (2) The actigraph readings were (significantly or nonsignificantly) negatively correlated with metabolic heat and nonsignificantly correlated with movement speeds. Hence, the actigraph validity for measuring PA intensity while strolling and jogging remains debatable.

show abstract

Section: Methodsologymentioning

confidence: 99%

Section: Literature Reviewmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Validity of Research-Grade Actigraphy Unit for Measuring Exercise Intensity

Han

Wang

2017

IJERPH

View full text Add to dashboard Cite

show abstract

“…Recent advances in commercial wearable technology has led to multiple devices that can enable PA to be assessed objectively. Of these, the accelerometer is commonly used for quantifying activity intensity and counting the number of steps [4]. Accelerometers are inexpensive, easy to use and longlasting.…”

Section: Introductionmentioning

confidence: 99%

Capturing accelerometer outputs in healthy volunteers under normal and simulated-pathological conditions using ML classifiers

Filippou

Redmond

Bennion

et al. 2020

2020 42nd Annual International Conference of the IEEE Engineering in Medicine &Amp; Biology Society (EMBC)

View full text Add to dashboard Cite

Wearable devices offer a possible solution for acquiring objective measurements of physical activity. Most current algorithms are derived using data from healthy volunteers. It is unclear whether such algorithms are suitable in specific clinical scenarios, such as when an individual has altered gait. We hypothesized that algorithms trained on healthy population will result in less accurate results when tested in individuals with altered gait. We further hypothesized that algorithms trained on simulated-pathological gait would prove better at classifying abnormal activity. We studied healthy volunteers to assess whether activity classification accuracy differed for those with healthy and simulated-pathological conditions. Healthy participants (n=30) were recruited from the University of Leeds to perform nine predefined activities under healthy and simulated-pathological conditions. Activities were captured using a wrist-worn MOX accelerometer (Maastricht Instruments, NL). Data were analyzed based on the Activity-Recognition-Chain process. We trained a Neural-Network, Random-Forests, k-Nearest-Neighbors (k-NN), Support-Vector-Machines (SVM) and Naive Bayes models to classify activity. Algorithms were trained four times; once with 'healthy' data, and once with 'simulatedpathological data' for each of activity-type and activity-task classification. In activity-type instances, the SVM provided the best results; the accuracy was 98.4% when the algorithm was trained and then tested with unseen data from the same group of healthy individuals. Accuracy dropped to 52.8% when tested on simulated-pathological data. When the model was retrained with simulated-pathological data, prediction accuracy for the corresponding test set was 96.7%. Algorithms developed on healthy data are less accurate for pathological conditions. When evaluating pathological conditions, classifier algorithms developed using data from a target sub-population can restore accuracy to above 95%. Clinical Relevance-This method remotely establishes healthrelated data of objective outcome measures of activities of daily living.

show abstract

“…Both these devices do have some form of computer with integrated sensors that can collect information relating to the user (Haghi, Thurow, & Stoll, 2017;Peltola, 2017). Due to their attentive and communicative qualities wearables devices have attracted attention from developers in various fields including health, fitness, sport, art and industrial applications (Walker, Hickey, & Freedson, 2016). Software and hardware developments in the wearables platform continue to grow due to the increased adoption of these devices for health monitoring purposes (Hutton et al, 2018).…”

Section: Introductionmentioning

confidence: 99%

Privacy in wearable health devices: How does POPIA measure up?

Katurura

Cilliers

Kalpa Publications in Computing

View full text Add to dashboard Cite

The market for wearable devices that is used for health monitoring has steadily increased over the past few years. South Africa has also seen an increase in the adoption of these wearable device. This is partly because these devices allow users to monitor their health and wellbeing in real time. However, to be efficient, the devices must collect a large amount of data. Some of the data that is collected include personally identifiable and health information which could be considered sensitive to the user. This study investigated if the Protection of Personal Information Act (POPIA) provides adequate protection to South African users of wearable health devices. The POPIA and the privacy policy of the 2 most popular wearable health devices in South Africa, the Apple watch and Fit- bit, were qualitatively compared making use of Hutton et al’s (2018) heuristic framework. The study found that POPIA protected the users’ privacy when it came to notice, users’ awareness, choice and consent, access and participation. The Act did not cover any privacy matters related to social disclosure of information by users. The study also found that Apple watch and Fitbit did well in protecting users’ privacy with regards to notice and awareness as well as access and participation. The two wearables performed poorest in regards to choice and consent as well as social disclosure controls to protect users’ privacy. The study recommend that users educate themselves in regard to how their data collected by wearable health devices are collected and protected.

show abstract

Advantages and Limitations of Wearable Activity Trackers: Considerations for Patients and Clinicians

Cited by 41 publications

References 30 publications

Validity of Research-Grade Actigraphy Unit for Measuring Exercise Intensity

Validity of Research-Grade Actigraphy Unit for Measuring Exercise Intensity

Capturing accelerometer outputs in healthy volunteers under normal and simulated-pathological conditions using ML classifiers

Privacy in wearable health devices: How does POPIA measure up?

Contact Info

Product

Resources

About