Content uploaded by André Henriksen
Author content
All content in this area was uploaded by André Henriksen on Nov 11, 2022
Content may be subject to copyright.
1/22
Collecting health-related research data using
consumer-based wireless smart scales
Erlend Johannessen1, Jonas Johansson2, Alexander Horsch1, Eirik Årsand1, Gunnar Hartvigsen1,3, André
Henriksen1
1Department of Computer Science, UiT, The Arctic University of Norway, Tromsø, Norway
2Department of Community Medicine, UiT, The Arctic University of Norway, Tromsø, Norway
3Department of Health and Nursing Science, University of Agder, Grimstad, Norway
Corresponding Author:
Erlend Johannessen, MSc (Comp Sci)
Department of Computer Science
University of Tromsø – The Arctic University of Norway
Postboks 6050 Langnes
Tromsø, 9037, Norway
Email: erlend.johannessen@uit.no
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4258684
Preprint not peer reviewed
2/22
Abstract
Background: Smart scales with wireless data transfer have existed for more than a decade. These smart
body composition scales can, together with smart watches and trackers, observe changes in the
population’s health. Combining body composition data with physical activity measurements from devices
such as smart watches could contribute to building a human digital twin.
Objective: The objectives of this study were to 1) investigate the evolution of smart scales in the last
decade, 2) map status and supported sensors of smart scales, 3) get an overview of how smart scales have
been used in research, and 4) identify smart scales for current and future research.
Method: We searched for devices through web shops and smart scale tests/reviews, extracting data from
the manufacturer’s official website, user manuals when available, and data from web shops. We also
searched scientific literature databases for smart scale usage in scientific papers.
Result: We identified 165 smart scales with a wireless connection from 72 different manufacturers,
released between 2009 and end of 2021. Of these devices, 49 (28%) had been discontinued by end of
2021.
Conclusion: The last six years have seen a distinct increase of these devices in the marketplace,
measuring body composition with bone mass, muscle mass, fat mass, and water mass, in addition to
weight. Still, the number of research projects featuring connected smart scales are few. One reason could
be the lack of professionally accurate measurements, though trend analysis might be a more feasible
usage scenario.
Keywords: Body composition, Body analysis, Connected smart scales, Data collection
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4258684
Preprint not peer reviewed
3/22
1 INTRODUCTION
Overweight and obesity are serious public-health concerns, which potentially can result in severe illnesses
such as diabetes, cancers, and cardiovascular diseases [1][2]. There is an escalating global epidemic of
overweight and obesity, and a substantial proportion of overweight and obesity cases are likely caused by
excess intake of food combined with a trend of decreases in physical activity. The World Health
Organization (WHO) has recently updated its guidelines for physical activity and sedentary behaviour [3],
in which WHO outlines recommended activity for children, adolescents and adults. Globally, 25% of
adults do not meet the recommended levels of physical activity [4].
Smart personal health devices can be used to monitor changes to population health because these devices
enable collecting continuous lifestyle data, potentially over a longer period compared to traditional
assessment methods. Smart watches and trackers will register biometric data such as activity intensity,
steps, and heart rate, while smart body composition scales can be used for measurements such as the
body’s water weight, muscle mass, bone mass, and visceral fat [5] [6] by using bio-electrical impedance
analysis (BIA) [7]. They may present us with a picture of our metabolism, skeletal health, and disease
risks. Combining data from smart wearables and smart scales can provide a more complete picture of
changes in the population’s health. This may reveal trends and shifts in population habits and create a
body of knowledge to assist regulating health policies and improve prevention and treatments procedures.
The goal of this study is to evaluate connected smart bodyweight scales for the consumer market, i.e.,
devices that can measure body-mass index (BMI), weight, and body composition. In addition, we aim to
identify brands that are used in research projects and consider which scales would be relevant for future
research in terms of data availability, sensor quality, and measurements.
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4258684
Preprint not peer reviewed
4/22
2 MATERIALS AND METHODS
We searched for the manufacturer’s self-reported data for technical data on the smart scales. To find
relevant smart scales, several avenues were explored:
1. Using Internet search to find smart scale reviews, and then using these to find smart scale
manufacturers [8][9][10][11].
2. Amazon bestseller lists to utilise the top 50 smart scales. For balance, bestseller lists for USA,
UK, Germany, India, and Japan were used, i.e., amazon.com, amazon.co.uk, amazon.de,
amazon.in, and amazon.jp, respectively.
3. Reviews from Henriksen et al. [12][13] about using fitness trackers and smart watches for
measuring physical activity as a starting point, since many manufacturers of wrist-worn devices
also make smart scales.
4. Smart scales supported by the Android-only openScale app [14], as used by the Quantified Self
community [15].
5. Recommendations from department colleagues.
The process further was to find each manufacturer’s website, to find the information on the individual
smart scales, and then complement the metadata from the Federal Communications Commission (FCC)
database [16], smart scale reviews, and web shops.
2.1 INCLUSION CRITERIA
Only data for smart scales with connectivity were retained since connectivity would facilitate self-
recording. Only consumer market smart scales were listed, since the professional smart scales would be
unavailable to, or too expensive for, the consumer market. If they were not targeted towards the consumer
market, they were not included. In addition, some of the scales were included but excluded for parts of the
analysis if the release year could not be identified.
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4258684
Preprint not peer reviewed
5/22
2.2 DATA COLLECTION
A total of 58 different variables were collected for each included weight scale, and Table 1 shows the
most relevant meta-data columns outside weight.
Variable
Description
Body fat
The total amount of body fat (aka. fat mass or fat percent)
Visceral fat
Fat surrounding the vital organs in the abdominal area
Subcutaneous fat
Fat tissue peripherally located throughout the body
Lean mass
Total body weight minus all body fat
Muscle mass
Muscle mass in the body
Protein
Fat-free mass minus water, minerals, and bone mass
Skeletal muscle
Muscle mass responsible for moving the body
Body water
The amount of body weight that is water
Bone mass
The amount of bone mineral content in the body
Table 1. Relevant variables collected for smart scales
Though not an exhaustive device search, the selected scales serve as a representative selection of current
smart scales. Data collection was done in 2021 between August 2nd and December 31st and contains
information on most smart scales available in this period.
Several manufacturers would not disclose all scale details on their website. We therefore had to collect
additional meta-data from reviewers and/or web shops. Using the Federal Communications Commission
(FCC) [17] database, some of the smart scales could also be found, with user manuals, reports, tests, and
specifications.
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4258684
Preprint not peer reviewed
6/22
2.3 SMART SCALES IN RESEARCH
A search phrase was set up for searching ACM Digital Library [18], IEEE Xplore [19], Ovid/MEDLINE
[20], PubMed [21], and Web of Science [22] for smart scales. The following criteria were used:
1) Must be a bodyweight type scale, used to weigh human adults.
2) Articles should state the model, or the brand of the smart scales used.
3) Must be available at the time of research, consumer-based smart scales, that anyone can buy in a
shop. I other words, they must not be professional-grade, or only available to general
practitioners, researchers, etc.
4) Must be able to send data to other devices or the internet through either Bluetooth, Wi-Fi or
cellular. If studies do not describe communication with other devices or the internet, a non-
connected bathroom scale is assumed, and article is excluded.
The search phrase used was ("bathroom scale" OR "weight scale" OR "e-scales" OR "smart scale" OR
"smart scales") AND ("body analysis" OR "body composition" OR "body weight" OR "body monitor" OR
BMI OR "body mass index") AND ("bluetooth" OR "wi-fi" OR wifi OR connected OR wireless), with an
exception for IEEE Xplore where a simpler version of the search was used. The query has three parts,
finding the correct type of scale, finding the correct type of functionality, and finding connected smart
scales.
3 RESULTS
3.1 RELEVANT SMART SCALES
The smart scale search procedure is presented with a flow-chart in Figure 1. We collected data for in total
181 smart scales from 72 manufacturers using the official manufacturer’s web sites, FCC database, web
shops or device reviews.
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4258684
Preprint not peer reviewed
7/22
We removed 10 scales because of lack of specifications, partial specification, or lack of reliable
information sources, and five devices for not being professional grade scales, i.e., not consumer based.
The remaining 165 devices were included in the study. For 28 devices release year was not available.
We also found that 28% of the remaining 165 devices were confirmed to be discontinued (n=47). This
number could be higher, though, given the lack of specific information from brand web sites.
Figure 1. A flow chart of the data collection process
3.1.1 Smart scale models by year
There has been an increasing trend of wireless smart scales since the first scale found from 2009. Year
2020 shows a peak in the set of scales we found, with 28 new scales released, while 2021 is lower with 14
new scales. The models of smart scales found for the period 2009-2015 were few, but from 2016 the
number of models increased. We detected a similar trend increase for new manufacturers.
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4258684
Preprint not peer reviewed
8/22
Figure 2. Smart scale models and manufacturers found, by release year (n=137)
Figure 2 shows the total number of smart scale models found each year. It also shows unique
manufacturers by year, i.e., counting one model per manufacturer, and new manufacturers, i.e.,
manufacturers that were not available the previous year.
Out of the included 165 smart scales, none had a cellular connection. The majority (84%) had Bluetooth
connectivity only, some had both Bluetooth and Wi-Fi connectivity (11%), and the remaining 5% were
Wi-Fi only.
3.2 TECHNOLOGY
All smart scales use weighing sensors, but in addition, BIA is used for measuring body composition. Of
the 165 selected scales, 153 had implemented BIA technology.
Two of the smart scales (InBody H20N and H20B [23]) also use a hand-held sensor “bar” while
measuring body composition, producing a more detailed measurement, which also give body composition
measurements for limbs vs. torso, e.g., how much fat in arms and legs vs. fat in torso. Thirty-two smart
scales had Indium tin oxide (ITO) surfaces, to increase measurement accuracy [24].
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4258684
Preprint not peer reviewed
9/22
Figure 3. Smart scale technology trends for the years 2015-2021
3.3 RELEVANT VARIABLES
Major variables for bio-electrical impedance analysis are fat percent, muscle mass, body water, and bone
mass. Visceral fat is also measured, but to a lesser degree. Figure 4 shows that lean mass, skeletal muscle,
protein mass, visceral fat and subcutaneous fat have increased since 2015, while BMI, body fat, body
water, muscle mass and bone mass have been stable around 80-90% in smart scales released.
Figure 4. Smart scale variable trends for the years 2015-2021
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4258684
Preprint not peer reviewed
10/22
3.4 USAGE IN RESEARCH
A total of 165 studies were found, where all research databases searched contributed items to the final list
of approved articles. After removing duplicates (n=54) and non-English items (n=2), 109 articles were
selected for screening. Irrelevant studies or studies not using connected scales (n=74) and professional
grade scales (n=8) were removed, leaving a total of 27 articles that met the inclusion criteria, see Figure 5.
Figure 5. PRISMA diagram for smart scale studies
The included studies can be divided into two groups, 1) data collection studies (n=20), and 2) validation
and analysis studies (n=7). One study referred to several smart scales, so the total number of smart scale
usage found within articles is n=29.
Withings smart scales were used in 13 studies [25]–[37], and Fitbit scales were used in seven studies
[38]–[44]. Brands found in other studies were A&D Medical [45][46], Philips [47], Renpho [48], Shenzen
Unique [49], Xioami [50], and Yunmai [51]. One study compared three smart scales, from Téfal,
Terraillon, and Withings with DEXA scan [25]. One study did not state a scale brand, and referred to the
scale as a “networked weight scale” [52].
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4258684
Preprint not peer reviewed
11/22
3.4.1 Connectivity
Of the 29 brands/models examined, 9 had Bluetooth only, 6 had Wi-Fi only, and 14 had both Bluetooth
and Wi-Fi. This also means that 23 of 29 (79%) scales had Bluetooth, and 20 out of 29 (69%) scales had
Wi-Fi. Cellular connection was consistently found to be on the professional scales only, and consequently
not included in the list of final studies.
4 DISCUSSION
4.1 SMART SCALE TRENDS
The earliest wireless smart scale found in this search was from 2009, the next one was in 2012, see Figure
2. Both were Withings scales. This may explain why Withings scales are the most prevalent in research
studies. The apparent lack of smart scales between 2009-2015 makes it difficult to find definitive trends
for this period, but more scales were found from 2016 so these form the basis of Figure 3 and Figure 4.
From 2013, and especially from 2016, the number of brands in the smart scale market seem to increase. In
2020, a there was a peak in the set of scales we found, while 2021 was notably lower. One can only
speculate if the COVID-19 pandemic influenced the drop in number of units. In addition, we cannot rule
out that some smart scales did not make it into our result list.
There still are some scales that do not have BIA. In our list only 12 smart scales (7.2%) lacked this
technology. Most notable is maybe the newest Fitbit smart scale from 2019, Fitbit Aria Air [53]. The
previous two Fitbit smart scales (Fitbit Aria, Fitbit Aria 2) both included body composition
measurements, while the more recent Fitbit Aria Air did not. This development may potentially come
from users not wanting or understanding body composition, or because this newer smart scale may give
better return of investment for the company. It remains to be seen if Fitbit in removing this option could
signify a trend in future smart scale development.
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4258684
Preprint not peer reviewed
12/22
Indium tin oxide surface covering was found on 19.3 percent of the selected smart scales. Looking at
Figure 3 we can see that ITO has been in use for many years and has had a minor increase in use since
2019. One concern with ITO is the cost of Indium and the need for higher temperature in the production
process [24], which may be a reason ITO is not more prevalent.
Weight, fat percent, muscle mass, body water, and bone mass have all largely been present in the
connected smart scales. These could be viewed as the “main” measurements. In addition, we have seen an
increasing trend for measurements like lean mass, skeletal muscle, protein mass, visceral fat, and
subcutaneous fat. There might be several reasons for this, such as better BIA accuracy being able to
produce more detailed parameters. Another reason may be that manufacturers add features to compete
with other brands, making their own scales more attractive for the consumer market.
4.2 SMART SCALE USAGE
Since scales are stationary and placed on the floor, the user needs to actively move to the scale, then
activate it and wait for the measurement to finish. This goes for weight measurement, but if the user in
addition is measuring body composition they need to remain still until the smart scale have indicated that
the measurement is completed. This could result in measurement errors or missing data if the user aborts
the weighing/measuring.
Smart scale connectivity type is another hurdle for uploading measurements. In case of a Bluetooth
connection, normally a mobile phone needs to be connected, usually with the mobile application active.
Some applications require the user to start the weighing from the mobile phone. All this makes it more
difficult for the user and might affect their motivation to complete the measurement. In the case of Wi-Fi
connection, the user must set up the Wi-Fi scale with network connection through Service Set Identifier
(SSID) and password, to be able to send measurements to the manufacturer’s cloud. This is still a hurdle,
but when it is done, the user only needs to stand still while measurements are taken.
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4258684
Preprint not peer reviewed
13/22
The by far least intrusive measurement setup is using a cellular connection. This uses mobile data
connection directly to send measurements to the manufacturer’s cloud. The user only has to place the
scale physically so that a connection to the mobile network is obtained. This suggests a reason for the
more professional smart scales to use this type of connection. The downside of this is that it is
subscription-based and adds expenses.
4.3 CONSUMER-BASED VS PROFESSIONAL DEVICES
Consumer-based smart scales differ from equipment used by health professionals, which may measure
single parameters only but with a much higher degree of accuracy [25]. The reliability of BIA in
consumer-based scales has been compared to professional body composition methods, and has been found
interchangeable on a population level [54]. Because some smart scale measurements may not be accurate
to a professional quality, the best way of using these is for the user to follow changes in measurements as
a trend, and not as an accurate measurement of true body tissue content.
4.4 MANUFACTURERS’ CLAIMED MEASUREMENTS
Smart scale manufacturers rely on capturing the public's interest to sell their scales. This implies
promoting ease of use, good design, accuracy, and reliability, but also measurements that potential users
find useful or interesting. However, there are limits to the accuracy of bioelectrical impedance
measurements [25][55][56]. Low priced smart scales usually mean limitations to accuracy, reliability, and
precision, especially in consumer-based smart scales. This is because manufacturers are not transparent in
how they measure or calculate the different variables, so results may deviate from the “true” value of a
body tissue. How much they deviate is not known since algorithms and measurement methods are not
disclosed by the manufacturers.
Measurements like subcutaneous and visceral fat are useful to quantify the proportion of harmful fat, but
less so if they are not accurate. Some of the attributes the manufacturers claim to measure are less
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4258684
Preprint not peer reviewed
14/22
plausible, especially because they do not declare their measurement techniques. One such measurement is
“protein” or “protein mass.” This is based on a calculation from lean mass minus water, minus minerals,
and is rarely, if ever, used in medical research the same way protein are described by manufacturers.
Manufacturers describe muscle mass, skeletal muscle, and protein as three different measurements,
although these are highly related to each other, and it is unlikely that they can be properly separated given
the proposed measurement techniques. Skeletal muscle is normally included in the term muscle mass, and
all muscle mass contains protein, though protein is also found throughout the body. One way to interpret
manufacturers' version of the term protein is "a number that can be easier for the user to keep tabs on
when following changes to the body based on nutrition intake and physical activity".
4.5 IMPLICATION FOR USAGE IN RESEARCH STUDIES
What may be considered when using connected smart scales in studies is device availability, data
obtainability, ease of use, and price. When gathering data from a population using a smart scale, ease of
use is paramount. If weighing becomes bothersome, study participants might stop using the scale, thereby
halting data measurement for the study. When a smart scale becomes “transparent,” in that the only thing
a user needs to do is step on it, wait, and then step off, it will be easier to use. This can be achieved by
using scales with Wi-Fi or cellular connection. The former needs Wi-Fi login, which could be more
technical. In contrast, the latter needs a cellular subscription, which is more expensive.
Two angles on smart scale studies are gathering data from people that already have smart scales at home
or supplying a population with smart scales to use. The first may skew the study population, the second is
expensive but may give data from a more heterogenous population. Also, studies that want the measured
data for subsequent processing should select smart scales from a manufacturer that has an API where
stored health data could be accessed.
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4258684
Preprint not peer reviewed
15/22
5 STRENGTHS AND LIMITATIONS
There are several limitations to this study. The manufacturers differed as to how much of the device
specification detail they would disclose, which means that there might be inaccuracies in the details for
some of the scales.
Metadata collection was done manually by one researcher, which limits how much data could be collected
due to time constraints. Collecting data manually may also be prone to human errors, but it may also be a
strength, in that potential errors also may be avoided that would surface in automatic data collection, e.g.,
web scraping [57].
Smart scales are available in different markets, Asia, the Americas, and Europe to name a few. Not all
these markets were thoroughly investigated, though top 50 smart scales in the Amazon web shop for
USA, UK, Germany, India, and Japan were examined to try to mitigate this.
6 CONCLUSIONS
This study focused specifically on consumer-based wireless smart scales because of the possibilities for
health-related data collection for a population.
Collecting data over time for an individual may provide health professionals with valuable information
regarding that person’s health status and may help inform prevention and treatment strategies for their
current health condition. Using a manufacturing term, this could be coined as a down-scaled “human
digital thread.” In our context it can be described as the data continuously generated by activity trackers
and smart scales.
Mair et.al. [58] propose using consumer-based activity trackers as data collection devices, with the caveat
that these devices have limitations. Consumer-based smart scales could also be added to the collection of
data collection devices.
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4258684
Preprint not peer reviewed
16/22
A combination of consumer-based activity trackers and smart scales would give a more comprehensive
insight into individuals’ health status. And even if individual measurements may not be precise for use in
health care directly, data collection for an entire population would even out the measurement
inconsistencies.
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4258684
Preprint not peer reviewed
17/22
Figures and Tables with Captions
Figure 1. A flow chart of the data collection process ...................................................................................7
Figure 2. Smart scale models and manufacturers found, by release year (n=137)........................................8
Figure 3. Smart scale technology trends for the years 2015-2021................................................................9
Figure 4. Smart scale variable trends for the years 2015-2021.....................................................................9
Figure 5. PRISMA diagram for smart scale studies....................................................................................10
Table 1. Relevant variables collected for smart scales ..................................................................................5
Author Contributions
EJ conceptualized and designed the study, with contributions from AH. EJ collected smart scales and
conducted the literature search for smart scale usage in research. EJ did the original draft preparation,
with critical review by all authors. All authors read and approved the final manuscript.
Funding Information
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Conflict of Interest
The authors declare that they have no known competing financial or commercial interests or personal
relationships that could have appeared to influence the work reported in this paper.
Acknowledgments
Thanks to Befolkingsundersøkelser i Nord (BiN) for support.
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4258684
Preprint not peer reviewed
18/22
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4258684
Preprint not peer reviewed
19/22
References
[1] G. M. Singh et al., ‘The Age-Specific Quantitative Effects of Metabolic Risk Factors on
Cardiovascular Diseases and Diabetes: A Pooled Analysis’, PLOS ONE, vol. 8, no. 7, p. e65174, Jul.
2013, doi: 10.1371/journal.pone.0065174.
[2] ‘World Obesity Day: “Missing the targets report”’, World Obesity Federation.
https://www.worldobesity.org/resources/resource-library/world-obesity-day-missing-the-targets-
report (accessed Jun. 13, 2022).
[3] ‘WHO guidelines on physical activity and sedentary behaviour’. https://www.who.int/publications-
detail-redirect/9789240015128 (accessed Nov. 01, 2021).
[4] ‘Physical activity’. https://www.who.int/news-room/fact-sheets/detail/physical-activity (accessed Jun.
14, 2022).
[5] T. Morgan, ‘Body composition explained’, Which? https://www.which.co.uk/reviews/bathroom-
scales/article/body-composition-explained-ahktC8j0kyDd (accessed Jun. 13, 2022).
[6] M. Kendall, ‘What Do the Numbers on My Bioimpedance Scale mean (and How Can I Manage
Them)?’, Optimising Nutrition, May 28, 2022. https://optimisingnutrition.com/what-do-the-numbers-
on-my-bioimpedance-scale-mean-and-how-can-i-manage-them/ (accessed May 31, 2022).
[7] R. N. Baumgartner, W. C. Chumlea, and A. F. Roche, ‘Estimation of body composition from
bioelectric impedance of body segments’, The American Journal of Clinical Nutrition, vol. 50, no. 2,
pp. 221–226, Aug. 1989, doi: 10.1093/ajcn/50.2.221.
[8] ‘Ranking The Best Bathroom Scales Of 2020 – Fitbug’. https://fitbug.com/bathroom-scale/ (accessed
Oct. 15, 2021).
[9] M. Howell, ‘9 of the best bathroom scales, including smart scales and body fat monitors’, The
Telegraph, Jun. 01, 2017. Accessed: Oct. 15, 2021. [Online]. Available:
https://www.telegraph.co.uk/health-fitness/body/best-bathroom-scales/
[10]T. E. Edwards Tanya, ‘The best smart scale of 2021’, CNN Underscored, Jan. 14, 2021.
https://www.cnn.com/2021/01/14/cnn-underscored/best-smart-scale/index.html (accessed Oct. 15,
2021).
[11]‘The Best Smart Bathroom Scales for 2021’, PCMAG. https://www.pcmag.com/picks/the-best-smart-
bathroom-scales (accessed Oct. 15, 2021).
[12]A. Henriksen et al., ‘Using Fitness Trackers and Smartwatches to Measure Physical Activity in
Research: Analysis of Consumer Wrist-Worn Wearables’, Journal of Medical Internet Research, vol.
20, p. e110, Mar. 2018, doi: 10.2196/jmir.9157.
[13]A. Henriksen, A. Z. Woldaregay, M. Muzny, G. Hartvigsen, L. A. Hopstock, and S. Grimsgaard,
‘Replication data for Using Fitness Trackers and Smartwatches to Measure Physical Activity in
Research’. DataverseNO, Oct. 29, 2021. doi: 10.18710/6ZWC9Z.
[14]OliE, openScale. 2021. Accessed: Oct. 19, 2021. [Online]. Available:
https://github.com/oliexdev/openScale
[15]‘Quantified Self Homepage’, Quantified Self. https://quantifiedself.com/ (accessed Oct. 19, 2021).
[16]F. ID, ‘FCC ID Search’, FCC ID. https://fccid.io/ (accessed Nov. 24, 2021).
[17]‘About the FCC’, Federal Communications Commission. https://www.fcc.gov/about/overview
(accessed Nov. 24, 2021).
[18]‘ACM Digital Library’. https://dl.acm.org/
[19]‘IEEE Xplore’. https://ieeexplore.ieee.org/ (accessed Feb. 04, 2022).
[20]‘Ovid: / MEDLINE’. https://ovidsp.dc1.ovid.com/ (accessed Feb. 04, 2022).
[21]‘PubMed’, PubMed. https://pubmed.ncbi.nlm.nih.gov/ (accessed Feb. 04, 2022).
[22]‘Web of Science’. https://www.webofscience.com/ (accessed Feb. 04, 2022).
[23]‘Smart Weight Full Body Composition Scale | InBody USA’, InBody.
https://consumer.inbodyusa.com/collections/h20-series (accessed Mar. 23, 2022).
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4258684
Preprint not peer reviewed
20/22
[24]E. B. Aydın and M. K. Sezgintürk, ‘Indium tin oxide (ITO): A promising material in biosensing
technology’, TrAC Trends in Analytical Chemistry, vol. 97, pp. 309–315, Dec. 2017, doi:
10.1016/j.trac.2017.09.021.
[25]J. Frija-Masson, J. Mullaert, E. Vidal-Petiot, N. Pons-Kerjean, M. Flamant, and M.-P. d’Ortho,
‘Accuracy of Smart Scales on Weight and Body Composition: Observational Study’, JMIR mHealth
and uHealth, vol. 9, no. 4, p. e22487, Apr. 2021, doi: 10.2196/22487.
[26]C.-C. Tsai, H.-K. Lu, L.-P. Chu, C.-C. Ting, K.-C. Hsieh, and C.-L. Lai, ‘Analysis of the Accuracy of
Resting Heart Rate with the Standing Electronic Weight Scale by Ballistocardiogram’, in 2018
International Symposium on Computer, Consumer and Control (IS3C), Dec. 2018, pp. 92–96. doi:
10.1109/IS3C.2018.00031.
[27]G. Wibisono and I. G. B. Astawa, ‘Designing Machine-to-Machine (M2M) Prototype System for
Weight Loss Program for Obesity and Overweight Patients’, in 2016 7th International Conference on
Intelligent Systems, Modelling and Simulation (ISMS), Jan. 2016, pp. 138–143. doi:
10.1109/ISMS.2016.52.
[28]P.-Y. Kuo and M. S. Horn, ‘Energy diet: energy feedback on a bathroom scale’, in Proceedings of the
2014 ACM International Joint Conference on Pervasive and Ubiquitous Computing, New York, NY,
USA, Sep. 2014, pp. 435–446. doi: 10.1145/2632048.2636063.
[29]J. M. Power, D. F. Tate, and C. G. Valle, ‘Experiences of African American Breast Cancer Survivors
Using Digital Scales and Activity Trackers in a Weight Gain Prevention Intervention: Qualitative
Study’, JMIR mHealth and uHealth, vol. 8, no. 6, p. e16059, Jun. 2020, doi: 10.2196/16059.
[30]A.-L. Vuorinen, E. Helander, J. Pietilä, and I. Korhonen, ‘Frequency of Self-Weighing and Weight
Change: Cohort Study With 10,000 Smart Scale Users’, Journal of Medical Internet Research, vol.
23, no. 6, p. e25529, Jun. 2021, doi: 10.2196/25529.
[31]D. Campo et al., ‘Measurement of Aortic Pulse Wave Velocity With a Connected Bathroom Scale’,
American Journal of Hypertension, vol. 30, no. 9, pp. 876–883, Sep. 2017, doi: 10.1093/ajh/hpx059.
[32]S. Mehrang, E. Helander, M. Pavel, A. Chieh, and I. Korhonen, ‘Outlier detection in weight time
series of connected scales’, in 2015 IEEE International Conference on Bioinformatics and
Biomedicine (BIBM), Nov. 2015, pp. 1489–1496. doi: 10.1109/BIBM.2015.7359896.
[33]C. G. Valle, A. M. Deal, and D. F. Tate, ‘Preventing weight gain in African American breast cancer
survivors using smart scales and activity trackers: a randomized controlled pilot study’, J Cancer
Surviv, vol. 11, no. 1, pp. 133–148, Feb. 2017, doi: 10.1007/s11764-016-0571-2.
[34]S. Mehrang, E. Helander, A. Chieh, and I. Korhonen, ‘Seasonal weight variation patterns in seven
countries located in northern and southern hemispheres’, in 2016 38th Annual International
Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Aug. 2016, pp.
2475–2478. doi: 10.1109/EMBC.2016.7591232.
[35]G. W. Kurtzman et al., ‘Social Incentives and Gamification to Promote Weight Loss: The LOSE IT
Randomized, Controlled Trial’, J GEN INTERN MED, vol. 33, no. 10, pp. 1669–1675, Oct. 2018,
doi: 10.1007/s11606-018-4552-1.
[36]L. Benson, F. Zhang, H. Espel-Huynh, L. Wilkinson, and M. R. Lowe, ‘Weight variability during
self-monitored weight loss predicts future weight loss outcome’, Int J Obes, vol. 44, no. 6, Art. no. 6,
Jun. 2020, doi: 10.1038/s41366-020-0534-6.
[37]M. Sperrin et al., ‘Who Self-Weighs and What Do They Gain From It? A Retrospective Comparison
Between Smart Scale Users and the General Population in England’, Journal of Medical Internet
Research, vol. 18, no. 1, p. e4767, Jan. 2016, doi: 10.2196/jmir.4767.
[38]L. Laranjo, J. C. Quiroz, H. L. Tong, M. A. Bazalar, and E. Coiera, ‘A Mobile Social Networking
App for Weight Management and Physical Activity Promotion: Results From an Experimental Mixed
Methods Study’, Journal of Medical Internet Research, vol. 22, no. 12, p. e19991, Dec. 2020, doi:
10.2196/19991.
[39]M. M. Marques et al., ‘A Theory- and Evidence-Based Digital Intervention Tool for Weight Loss
Maintenance (NoHoW Toolkit): Systematic Development and Refinement Study’, Journal of
Medical Internet Research, vol. 23, no. 12, p. e25305, Dec. 2021, doi: 10.2196/25305.
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4258684
Preprint not peer reviewed
21/22
[40]G. Wibisono and I. G. B. Astawa, ‘Designing Machine-to-Machine (M2M) Prototype System for
Weight Loss Program for Obesity and Overweight Patients’, in 2016 7th International Conference on
Intelligent Systems, Modelling and Simulation (ISMS), Jan. 2016, pp. 138–143. doi:
10.1109/ISMS.2016.52.
[41]J. Turicchi et al., ‘Body weight variability is not associated with changes in risk factors for
cardiometabolic disease’, International Journal of Cardiology Hypertension, vol. 6, p. 100045, Sep.
2020, doi: 10.1016/j.ijchy.2020.100045.
[42]J. Turicchi et al., ‘Data Imputation and Body Weight Variability Calculation Using Linear and
Nonlinear Methods in Data Collected From Digital Smart Scales: Simulation and Validation Study’,
JMIR mHealth and uHealth, vol. 8, no. 9, p. e17977, Sep. 2020, doi: 10.2196/17977.
[43]J. Turicchi et al., ‘The impact of early body-weight variability on long-term weight maintenance:
exploratory results from the NoHoW weight-loss maintenance intervention’, Int J Obes, vol. 45, no.
3, Art. no. 3, Mar. 2021, doi: 10.1038/s41366-020-00706-0.
[44]K. M. Hood et al., ‘Validity and reliability of a Wi-Fi smart scale to estimate body composition’,
Health Technol., vol. 9, no. 5, pp. 839–846, Nov. 2019, doi: 10.1007/s12553-019-00359-8.
[45]N. Nakashima, T. Hiramatsu, P. P. Ghosh, R. Islam, K. Kobayashi, and T. Inoguchi, ‘Evaluation of
“Portable Health Clinic” with BAN standard for 10K subjects in Bangladesh’, in 2013 35th Annual
International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Jul.
2013, pp. 1672–1675. doi: 10.1109/EMBC.2013.6609839.
[46]‘Remote Health Monitoring for Older Adults and Those with Heart Failure: Adherence and System
Usability | Telemedicine and e-Health’. https://www.liebertpub.com/doi/10.1089/tmj.2015.0140
(accessed Feb. 09, 2022).
[47]I. C. Gyllensten et al., ‘Early Indication of Decompensated Heart Failure in Patients on Home-
Telemonitoring: A Comparison of Prediction Algorithms Based on Daily Weight and Noninvasive
Transthoracic Bio-impedance’, JMIR Medical Informatics, vol. 4, no. 1, p. e4842, Feb. 2016, doi:
10.2196/medinform.4842.
[48]L. Cordero et al., ‘MO663THE UTILITY OF SMART SCALES TO MONITOR BODY
COMPOSITION IN HEMODIALYSIS PATIENTS’, Nephrology Dialysis Transplantation, vol. 36,
no. Supplement_1, p. gfab099.008, May 2021, doi: 10.1093/ndt/gfab099.008.
[49]E. Everett, B. Kane, A. Yoo, A. Dobs, and N. Mathioudakis, ‘A Novel Approach for Fully
Automated, Personalized Health Coaching for Adults with Prediabetes: Pilot Clinical Trial’, Journal
of Medical Internet Research, vol. 20, no. 2, p. e9723, Feb. 2018, doi: 10.2196/jmir.9723.
[50]‘Suspension training classes as an effective means to reduce the bodyfat level in middle-aged women
- Minerva Ortopedica e Traumatologica 2020 June;71(2):61-7’.
https://www.minervamedica.it/en/journals/minerva-
orthopedics/article.php?cod=R14Y2020N02A0061 (accessed Feb. 09, 2022).
[51]M. L. Butryn, K. M. Godfrey, M. K. Martinelli, S. R. Roberts, E. M. Forman, and F. Zhang, ‘Digital
self-monitoring: Does adherence or association with outcomes differ by self-monitoring target?’,
Obesity Science & Practice, vol. 6, no. 2, pp. 126–133, 2020, doi: 10.1002/osp4.391.
[52]F. Chen et al., ‘Clinical and Economic Impact of a Digital, Remotely-Delivered Intensive Behavioral
Counseling Program on Medicare Beneficiaries at Risk for Diabetes and Cardiovascular Disease’,
PLOS ONE, vol. 11, no. 10, p. e0163627, Oct. 2016, doi: 10.1371/journal.pone.0163627.
[53]‘Fitbit Aria Air Smart Scales | Shop’. https://www.fitbit.com/global/us/products/scales/aria-air
(accessed Feb. 10, 2022).
[54]N. Achamrah et al., ‘Comparison of body composition assessment by DXA and BIA according to the
body mass index: A retrospective study on 3655 measures’, PLOS ONE, vol. 13, no. 7, p. e0200465,
Jul. 2018, doi: 10.1371/journal.pone.0200465.
[55]J. T. Peterson, W. E. Repovich, M. Eash, D. Notrica, and C. R. Hill, ‘Accuracy of Consumer Grade
Bioelectrical Impedance Analysis Devices Compared to Air Displacement Plethysmography: 2105:
Board #18: June 1 8:00 AM - 9:30 AM’, Medicine & Science in Sports & Exercise, vol. 39, no. 5, p.
S373, May 2007, doi: 10.1249/01.mss.0000274464.50186.ed.
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4258684
Preprint not peer reviewed
22/22
[56]J. Barnas, N. Micinski, L. Wheatley, K. Laurson, D. Thomas, and D. Brown, ‘Accuracy of Consumer
Grade Bioelectrical Impedance Analyzers to Assess Body Composition.: 2319 Board #24 May 30, 9:
30 AM - 11: 00 AM’, Medicine & Science in Sports & Exercise, vol. 46, no. 5S, p. 619, May 2014,
doi: 10.1249/01.mss.0000495325.84960.c3.
[57]‘7 Web Scraping Limitations You Should Know’. https://www.octoparse.com/blog/web-scraping-
limitations (accessed Oct. 11, 2022).
[58]J. L. Mair, L. D. Hayes, A. K. Campbell, and N. Sculthorpe, ‘Should We Use Activity Tracker Data
From Smartphones and Wearables to Understand Population Physical Activity Patterns?’, Journal for
the Measurement of Physical Behaviour, vol. 1, no. aop, pp. 1–5, Oct. 2021, doi: 10.1123/jmpb.2021-
0012.
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4258684
Preprint not peer reviewed
