CoChat
Sign Up
Cited by 451Context-Aware Activity Recognition SystemsNon-Invasive Vital Sign MonitoringWireless Sensor Networks for Data Analysis

Review of Wearable Devices and Data Collection Considerations for Connected Health

Open access
Vijayan, Vini, Connolly, James P., Condell, Joan, McKelvey, Nigel, Gardiner, Philip·2021·Sensors
DOI 10.3390/s21165589PMC PMC8402237
Open this paper in CoChat — free accountOpen PDF

Abstract

Wearable sensor technology has gradually extended its usability into a wide range of well-known applications. Wearable sensors can typically assess and quantify the wearer’s physiology and are commonly employed for human activity detection and quantified self-assessment. Wearable sensors are increasingly utilised to monitor patient health, rapidly assist with disease diagnosis, and help predict and often improve patient outcomes. Clinicians use various self-report questionnaires and well-known tests to report patient symptoms and assess their functional ability. These assessments are time consuming and costly and depend on subjective patient recall. Moreover, measurements may not accurately demonstrate the patient’s functional ability whilst at home. Wearable sensors can be used to detect and quantify specific movements in different applications. The volume of data collected by wearable sensors during long-term assessment of ambulatory movement can become immense in tuple size. This paper discusses current techniques used to track and record various human body movements, as well as techniques used to measure activity and sleep from long-term data collected by wearable technology devices.

Verified 2026-10-09 · 1 reference not checked yet

AI snapshot

Written from the paper’s full text.

TL;DR

This review surveys wearable devices and data-collection considerations for connected health, covering clinical and ambulatory monitoring, quantified self, measurement accuracy, privacy/security, and deep-learning methods for activity and sleep recognition. It concludes that wearables can provide objective long-term data, but accuracy, validation, power, connectivity, and security remain key challenges.

Key findings

  • Wearable sensors are used for activity, sleep, rehabilitation, chronic disease, and musculoskeletal assessment; global connected wearable devices were approximately 600 million in 2020, predicted to rise to 928 million in 2021 and 1100 million in 2022.
  • IMUs are reported to outperform other hardware sensors regarding measurement accuracy across applications; wearable device accuracy and precision typically range from 92% to 99%, and many smartwatches/smartphones report >95% step-count accuracy.
  • Reviewed HAR/DL studies report varied results, including CNN accuracy of 84%, U-Net up to 94.7% accuracy (table reports max accuracy 96.4% and Fw-score 0.965), InnoHAR F-scores of 0.946, 0.935, and 0.945 across three datasets, WaistonBelt X F1 score of 0.82, and OpenHealth gesture recognition accuracy of 98.6%.
  • Sleep/activity algorithms show mixed performance: GGIR classified seven activity types with 91% accuracy for hip-worn and 88% for wrist-worn accelerometers in 52 children; Cole–Kripke sleep/wake state was 96.13% effective in 49 adults; in-home MMWA using Cole–Kripke showed 94–98% sensitivity and 42–54% specificity for sleep epochs.

Methods

  • Narrative review of wearable devices and assessment techniques used in self-care and healthcare, including clinical trials, quantified self, measurement accuracy, privacy/security, and deep-learning approaches for activity and sleep identification.
  • Synthesizes literature and tables comparing wearable devices, sensor technologies, mean accuracy, smartwatches/activity trackers, and ML/NN models with participant counts and reported results.
  • No primary data collection or systematic review protocol is reported in the text; evidence is drawn from cited studies and device examples.

Limitations

  • Stated: wearable device accuracy and validity remain difficult challenges; main assessment problems include power consumption, connectivity capability, design constraints, and security concerns; data privacy and security remain unresolved, and many devices store data locally without encryption or data protection.
  • Stated: measurement accuracy in patients with low activity is often lacking; step count and activity determination are often unreliable when gait speed is slow and/or uneven; Cole–Kripke sleep misclassification was high in children compared with adults; GGIR and Cole–Kripke do not provide facilities to segment data based on specific movement patterns.
  • Evident: as a narrative review, it reports no formal search strategy, inclusion criteria, or meta-analysis, and the cited accuracy figures come from heterogeneous devices, populations, and protocols, limiting direct comparison.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.3390/s21165589

Cites

146 of 148 references verified0 citation errors · 1 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 116 cited sources checked · 0 supported · 0 flagged

  • Premisecitation

    Wearable technology decreases the cost of intensive treatment by allowing rehabilitation outside the hospital in an ambulatory environment.

    “Furthermore, wearable technology decreases the cost of intensive treatment by allowing rehabilitation outside of the hospital in an ambulatory environment [3].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [3] Frost, 2014, Market Research Report, Strategy R

      Not checked in the free run.

  • Premisecitation

    Wearable sensors are widely used in healthcare because of their hardware capacity, small footprint, and lower cost compared with equivalent medical instruments capable of monitoring the same vital signs.

    “Wearable sensors are widely used in healthcare, due to their hardware capacity, small footprint and lower cost compared to equivalent medical instruments capable of monitoring the same vital signs [2].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [2] Liu, S. (2021, August 03). Wearable Technology–Statistics & Facts Statista. Available online: www.statista.com/topics/1556/wearable-technology/.

      Not checked in the free run.

  • Premisecitation

    Data collected from each sensor can be used as additional information to further inform a clinician during a patient health status assessment of improvement or decline.

    “Data collected from each sensor can then be used as additional information to further inform a clinician during a patient health status assessment of improvement or decline [24].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [24] Dhairya, K. (2021, August 03). Introduction to Data Preprocessing in Machine Learning. Available online: https://towardsdatascience.com/introduction-to-data-preprocessing-in-machine-learning-a9fa83a5dc9.

      Not checked in the free run.

  • Premisecitation

    Wearable devices transfer captured data to a powerful remote computer or a cloud implementation, where the sensor information is deciphered, deconstructed, and results are meaningfully generated, interpreted and presented to the user.

    “As a result, they transfer captured data to a powerful remote computer or a cloud implementation, where the sensor information is deciphered, deconstructed, and results are meaningfully generated, interpreted and presented to the user [12].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [12] Majumder, S., Mondal, T., and Deen, M.J. (2017). Wearable Sensors for Remote Health Monitoring. Sensors, 17. source

      Not checked in the free run.

  • Premisecitation

    Wearable sensors are commonly used to assist in the transition of patient treatment from the clinic to the ambulatory environment.

    “Wearable sensors are commonly used to assist in the transition of patient treatment from the clinic to the ambulatory environment.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [20] Bell (2019). J. Wearable Health Monitoring Systems, Technical Report; Nyx Illuminated Clothing Company.

      Not checked in the free run.

  • Classified as probably not a substantial claim. 40
  • May combine two claims. 38
  • The quote only partly states this claim. 2

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Lightweight wireless pulse oximeters that detect oxygen levels in the blood by analysing real-time patient data are commonly used with COPD.

    “Lightweight wireless pulse oximeters that detect oxygen levels in the blood by analysing real-time patient data are commonly used with COPD [8,9].”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [8] Massaroni, 2015, Medical Smart Textiles Based on Fiber Optic Technology: An Overview, J. Funct. Biomater. source

      Not checked in the free run.

    • cites [9] Jouffroy, 2020, Prehospital pulse oximetry: A red flag for early detection of silent hypoxemia in COVID-19 patients, Crit. Care source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    The attachable skin format of these sensors improves the portability of otherwise cumbersome devices and can optimise bioelectrical signals obtained from users.

    “Additionally, the attachable skin format of these sensors improves the portability of the otherwise cumbersome devices, and can optimise bioelectrical signals obtained from users [7].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [7] Song, M.-S., Kang, S.-G., Lee, K.-T., and Kim, J. (2019). Wireless, Skin-Mountable EMG Sensor for Human–Machine Interface Application. Micromachines, 10. source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Wearable devices can inter-communicate using the human body as a transmission channel or through an appropriate transmission medium such as BLE, Zigbee or Wi-Fi.

    “They can inter-communicate using the human body as a transmission channel [11] or through an appropriate transmission medium, such as BLE, Zigbee or Wi-Fi.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [11] Vijayan, V., McKelvey, N., Condell, J., Gardiner, P., and Connolly, J. (2020, January 11–12). Implementing Pattern Recognition and Matching techniques to automatically detect standardized functional tests from wearable technology. Proceedings of the 2020 31st Irish Signals and Systems Conference (ISSC), Letterkenny, Ireland. source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Wearable devices can provide real-time feedback regarding a person's health conditions and thus provide an objective alternative to manage and monitor chronic disease progression in the elderly, within rehabilitation, and for those with various disabilities.

    “Wearable devices can provide real-time feedback regarding a person’s health conditions; hence, they can provide an objective alternative to manage and monitor chronic disease progression, such as with the elderly, within rehabilitation, and for those with various disabilities [1].”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [1] Pantelopoulos, 2009, A Survey on Wearable Sensor-Based Systems for Health Monitoring and Prognosis, IEEE Trans. Syst. Man. Cybern. Part C Appl. Rev. source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Optical sensors that use a light-based technique to quantify the delicate magnetic fields produced by neurons firing may be used instead of MRI machines to create similar imaging, eliminating the expensive cooling or electromagnetic shielding required for MRI scans.

    “Optical sensors that use a light-based technique to quantify the delicate magnetic fields produced by neurons firing in the brain may be used instead of MRI machines to create similar imaging, eliminating the expensive cooling or electromagnetic shielding required when patients require an MRI scan [4].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [4] Pando, A. (2021, May 31). Wearable Health Technologies and Their Impact on the Health Industry. Forbes. Available online: www.forbes.com/sites/forbestechcouncil/2019/05/02/wearable-health-technologies-and-their-impact-on-the-health-industry/#4eac185f3af5.

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Wearables have been used in the UK to remotely monitor people with chronic disease or post-covid symptoms, and clinicians can remotely examine, regularly view vital sign data and provide adequate remote consultations.

    “Wearables have been used in the UK to remotely monitor people with chronic disease or post-covid symptoms, and clinicians can remotely examine, regularly view vital sign data and provide adequate remote consultations [10].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [10] Best, 2021, Wearable technology: Covid-19 and the rise of remote clinical monitoring, BMJ source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

    Premisecitation

    Before a wearable device can be used for clinical or research purposes, it is important to understand the different types of data that may be required throughout the duration from a data collection perspective.

    “Before a wearable device can be used for clinical or research purposes, it is important to understand the different types of data that may be required throughout the duration from a data collection perspective [22].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [22] Hurdles, C. (2021, August 03). Applied Clinical Trials- Your Peer-Reviewed Guide to Gobal Clinical Trials Management. Available online: https://cdn.sanity.io/files/0vv8moc6/act/346a82766960a17da2099b5f1268d0efba485d2a.pdf.

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Automatic human activity recognition is an essential element in developing human-interaction applications in personal fitness and healthcare.

    “Automatic human activity recognition is an essential element in developing human-interaction applications in personal fitness and healthcare [14].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [14] Schrader, 2020, Advanced Sensing and Human Activity Recognition in Early Intervention and Rehabilitation of Elderly People, J. Popul. Ageing source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Traditional clinical trial questionnaires and tests are used by physicians for musculoskeletal assessments such as Rheumatoid Arthritis (RA) and Ankylosing Spondylitis (AS).

    “These traditional clinical trial questionnaires and tests are used by physicians for musculoskeletal assessments, such as Rheumatoid Arthritis (RA) and Ankylosing Spondylitis (AS) [26].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [26] Balsa, 2004, Value of Disease Activity Score 28 (DAS28) and DAS28-3 Compared to American College of Rheumatology-Defined Remission in Rheumatoid Arthritis, J. Rheumatol.

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Fifth generation (5G) communication technology improves broadband networks to support always-available transmission services, low end-to-end latency, and a significant data rate boost over 4G.

    “Fifth generation (5G) communication technology improves broadband networks to support demanding requirements, such as always-available transmission services, low end-to-end latency, and a significant data rate boost over fourth generation (4G) communication technologies [13].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [13] Sun, 2018, Wearable Communications in 5G: Challenges and Enabling Technologies, IEEE Veh. Technol. Mag. source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    An IMU device contains an accelerometer, gyroscope and magnetometer that are combined to create independent measurements of the locations of an object to which it is attached.

    “An IMU device contains an accelerometer, gyroscope and magnetometer that are combined to create independent measurements of the locations of an object to which it is attached [27].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [27] Callmer, J. (2016). Autonomous Localization in Unknown Environments. [Master’s Thesis, Linköping University Electronic Press].

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Inertial Measurement Unit (IMU) sensors are now widely deployed in professional sports.

    “Athletes have several specific sensor-based applications, and Inertial Measurement Unit (IMU) sensors are now widely deployed in professional sports [15].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [15] Cha, J., Kim, J., and Kim, S. (2019). Hands-free user interface for AR/VR devices exploiting wearer’s facial gestures using unsupervised deep learning. Sensors, 19. source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    The ability to accurately quantify movement dates back around 30 years with developing motion capture gait laboratories based upon techniques such as optical photogrammetry in Vicon, magnetometry such as Fastrack and force platform walkways including GaitRite.

    “The ability to accurately quantify movement dates back around 30 years with developing motion capture gait laboratories based upon techniques, such as optical photogrammetry in Vicon, magnetometry, such as Fastrack and force platform walkways, including GaitRite [86].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [86] Wilson, S., and Laing, R.M. (2018, January 23–26). Wearable Technology: Present and Future. Proceedings of the 91st World Conference, Leeds, UK.

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Wearable sensors can accurately determine the angle of joint motion, neck movement, head rotation and flexion and extension movement.

    “These sensors can accurately determine the angle of joint motion, neck movement, head rotation and flexion and extension movement [38].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [38] Lawton, 1969, ADL and IADL treatment; Assessment of older people: Self-maintaining and instrumental activities of daily living, Gerontologist source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    If given correct cut-off frequencies, the predicted output of a complementary filter is practically noise-free and should enable quality data to flow through the filter.

    “If given with correct cut-off frequencies, the predicted output of a complementary filter is practically noise-free and should enable quality data to flow through the filter [35].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [35] Paina, G.P., Gaydou, D., Redolfi, J., Paz, C., and Canali, L. (2011). Experimental comparison of kalman and complementary filter for attitude estimation. Proc. AST, 205–215.

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Premisecitation

    A smartphone program can accurately predict RA disease activity based on a non-invasive self-assessment of a combination of disabilities in everyday tasks, joint symptoms and walking ability.

    “It proposes that a smartphone program can accurately predict RA disease activity based on a non-invasive self-assessment of a combination of disabilities in everyday tasks, joint symptoms and walking ability [63].”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [63] Nishiguchi, 2014, Self-Assessment Tool of Disease Activity of Rheumatoid Arthritis by Using a Smartphone Application, Telemed. e-Health source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Data generated by an IMU may be incomplete, imprecise, or prone to error because accelerometer data may be tainted by gravitational force, nearby magnetic disturbances may affect magnetometer measurements, and gyroscope data may drift due to unbounded time-varying factors.

    “Data generated by an IMU may be incomplete, imprecise, or prone to error, due to (i) accelerometer data may be tainted by a gravitational force experienced by the IMU device [29], (ii) nearby magnetic disturbances may affect magnetometer measurements, and (iii) gyroscope data may drift due to unbounded time-varying factors [30].”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [29] Pandey, 2008, Transverse vibrations driven negative thermal expansion in a metallic compound GdPd3B0.25C0.75, Appl. Phys. Lett. source

      Not checked in the free run.

    • cites [30] Sparkfun (2021, August 03). Accelerometer, Gyro and IMU Buying Guide–SparkFun Electronics. Available online: www.sparkfun.com/pages/accel_gyro_guide.

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    The vulnerability of wireless communication channels can result in data loss during transmission, as well as errors in data analysis and prediction.

    “The vulnerability of wireless communication channels can result in data loss during transmission, as well as errors in data analysis and prediction [109].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [109] Yu, 2015, A Study on Data Loss Compensation of WiFi-Based Wireless Sensor Networks for Structural Health Monitoring, IEEE Sensors J. source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Sensor-intrinsic noise is typically caused by resistive sensor temperature noise and repetitive noise in capacitive sensors.

    “Sensor-intrinsic noise is typically caused by resistive sensor temperature noise and repetitive noise in capacitive sensors [108].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [108] Nagel, 2009, Noise in MEMS, Meas. Sci. Technol.

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Technical requirements for wearable technology in healthcare applications include that they are fit-for-purpose for various use cases within the specific domain, can collect data in real-life environments, and are precise, consistent and tested in the healthcare community in which they would be utilised.

    “Technical requirements for wearable technology in healthcare applications include that they are fit-for-purpose for various use cases within the specific domain, can collect data in real-life environments, and are precise, consistent and tested in the healthcare community in which they would be utilised [94].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [94] Tim, 2018, Getting the Most out of Wearable Technology in Clinical Research, J. Clin. Stud.

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Motion artefact is commonly caused by sensor movement over the skin, temperature differences at the start and end of physical exercises, blood flow patterns, surface structure and thickness of the skin, and tissue density under the skin.

    “It is commonly caused by sensor movement over the skin, temperature differences at the start and end of physical exercises, blood flow patterns, surface structure and thickness of the skin, and tissue density under the skin [98].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [98] Bent, 2020, Investigating sources of inaccuracy in wearable optical heart rate sensors, NPJ Digit. Med. source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Formal gait laboratory kinematic assessments are more accurate than other assessments but are expensive and their availability is limited to a small number of specialist centres.

    “Formal gait laboratory kinematic assessments are more accurate, but expensive, and availability is limited to a small number of specialist centres [88].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [88] Maenner, 2013, Evaluation of an activities of daily living scale for adolescents and adults with developmental disabilities, Disabil. Heal. J. source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    For wearable devices with cameras and motion sensors used in rehabilitation exercise evaluation for knee osteoarthritis, the accuracy for activity type categorisation was 97%, while the accuracy for exercise posture recognition was 88%.

    “The accuracy for activity type categorisation was 97%, while the accuracy for exercise posture recognition was 88% [91].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [91] Chen, 2015, Wearable Sensor-Based Rehabilitation Exercise Assessment for Knee Osteoarthritis, Sensors source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    A high sampling rate contains more data than a low sampling rate, resulting in a hysteresis effect in the overall output.

    “It is obvious that a high sampling rate contains more data than a low sampling rate, resulting in a hysteresis effect in the overall output [115].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [115] Bieber, G., Haescher, M., and Vahl, M. (2013, January 29–31). Sensor requirements for activity recognition on smart watches. Proceedings of the 6th International Conference on PErvasive Technologies Related to Assistive Environments, Rhodes, Greece. source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Many wearable sensors, such as headbands and textile sensors, are effective in the short-term for activity tracking inside a controlled environment, but are inaccurate in the long-term.

    “Many wearable sensors, such as headbands and textile sensors, are effective in the short-term for activity tracking inside a controlled environment, but are inaccurate in the long-term [8].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [8] Massaroni, 2015, Medical Smart Textiles Based on Fiber Optic Technology: An Overview, J. Funct. Biomater. source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Wearable devices inspire people to work out more, but they may leave them frustrated if they fail to meet the recommended activity level.

    “Wearable devices inspire people to work out more, but they may leave them frustrated if they fail to meet the recommended activity level [123].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [123] Healthcare-in-Europe (2021, February 23). Smart Watches and Fitness Trackers Useful but May Increase Anxiety. Available online: https://healthcare-in-europe.com/en/news/smart-watches-fitness-trackers-useful-but-may-increase-anxiety.html.

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    A survey by the Institute of Health Matrix and Evaluation (IHME) estimated that more than one of every six individuals in European countries have a mental health problem.

    “A survey by the Institute of Health Matrix and Evaluation (IHME) estimated that more than one of every six individuals in European countries have a mental health problem [128].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [128] (2021, May 10). Projects Institute for Health Metrics and Evaluation. Available online: www.healthdata.org/projects.

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Epidermal biosensors such as a graphene electronic tattoo (GET) resemble a tattoo and are not distinguishable when worn, thus providing more confidence in using this type of wearable sensor to capture disease data.

    “This type of sensor resembles a tattoo, and is not distinguishable when worn, thus providing more confidence in using this type of wearable sensor to capture disease data [121].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [121] Ameri, 2017, Graphene Electronic Tattoo Sensors, ACS Nano source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    A Fitbit device was subjected to data injection attacks, battery drain hacks and denial-of-service attacks.

    “Recent research indicated how a Fitbit device was subjected to data injection attacks, battery drain hacks and denial-of-service attacks [130].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [130] Tawalbeh, L., Muheidat, F., Tawalbeh, M., and Quwaider, M. (2020). IoT Privacy and Security: Challenges and Solutions. Appl. Sci., 10. source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Studies show that those who use quantified-self applications feel more in control over their mood, which helps them control their mood and show more confidence, a positive attitude and a better outlook towards their emotional wellbeing.

    “Studies [126,127] show that those who use QS applications feel more in control over their mood, which helps them control their mood and show more confidence, a positive attitude and a better outlook towards their emotional wellbeing.”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [126] Mendu, S., and Baee, S. (2020). Redesigning the Quantified Self Ecosystem with Mental Health in Mind, ACM.

      Not checked in the free run.

    • cites [127] Majumdar, N. (2021, February 05). Quantified Self Detecting and Resolving Depression by Your Mobile Phone–Emberify Blog. Available online: https://emberify.com/blog/quantified-self-depression/.

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Larger batteries tend to reduce charging time and increase between-charging time, but they are uncomfortable to use because they are larger in form size than smaller batteries and increase the overall dimensions of the wearable device.

    “Larger batteries tend to reduce charging time and increase between-charging time, but they are uncomfortable to use, since they are larger in form size than smaller batteries and increase the overall dimensions of the wearable device [122].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [122] Chandel, V., Sinharay, A., Ahmed, N., and Ghose, A. (2016, January 30). Exploiting IMU Sensors for IOT Enabled Health Monitoring. Proceedings of the First Workshop on IoT-Enabled Healthcare and Wellness Technologies and Systems, Singapore. source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    System training for semi-supervised learning combines small quantities of a fully labelled dataset with large quantities of unlabelled data.

    “System training for semi-supervised learning system combines small quantities of a fully labelled dataset with large quantities of unlabelled data [147].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [147] Hailat, Z., Komarichev, A., and Chen, X.-W. (2018, January 20–24). Deep Semi-Supervised Learning. Proceedings of the 2018 24th International Conference on Pattern Recognition (ICPR), Beijing, China. source

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Interpretationreasoning

    Data privacy and security are critical in healthcare applications that use wearable devices because health information is sensitive or confidential.

    “Since health information is sensitive or confidential in nature, data privacy and security are critical in healthcare applications that use wearable devices.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Health information is sensitive or confidential in nature.

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

    Interpretationnone given

    Diligent use of wearable devices limits the likelihood of data loss.

    “However, diligent use of these devices limits the likelihood of data loss.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Supervised learning algorithms are trained by fully labelled datasets and can then make predictions on new data.

    “Supervised learning algorithms are trained by fully labelled datasets [140], and can then make predictions on new data [141].”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [140] Gmyzin, D. (2017). A Comparison of Supervised Machine Learning Classification Techniques and Theory-Driven Approaches for the Prediction of Subjective Mental Workload Subjective Mental. [Master’s Thesis, Technological University Dublin].

      Not checked in the free run.

    • cites [141] Osisanwo, 2017, Supervised Machine Learning Algorithms: Classification and Comparison, Int. J. Comput. Trends Technol. source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Unsupervised learning algorithms process unlabelled data and find patterns and relationships from them.

    “Unsupervised learning algorithms process unlabelled data and find patterns and relationships from them [145].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [145] Ben-gal, I. (2014). Outlier detection Irad Ben-Gal Department of Industrial Engineering. Data Mining and Knowledge Discovery Handbook, Springer.

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based).

    Interpretationnone given

    It is impractical to use wearable devices with complete data protection.

    “It is impractical to use wearable devices with complete data protection.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

    Interpretationnone given

    Continuous human movement detection and analysis is a tough problem in machine learning, since it is difficult to define the characteristics and duration of various movement patterns.

    “Continuous human movement detection and analysis is a tough problem in ML, since it is difficult to define the characteristics and duration of various movement patterns.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: Human motion characteristics vary from person to person.
    • Rests on an assumption: Each person’s motions of body components, such as the arm, leg and head for regular activities, such as walking, jogging and eating, are unique.
    • Rests on an assumption: The time required to do each of these activities also varies between individuals.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Deep learning is a subset of machine learning which can learn from unsupervised data and provides a system to learn from those datasets.

    “DL is a subset of ML which can learn from unsupervised data and provides a system to learn from those datasets [149].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [149] Lee, J., Bahri, Y., Novak, R., Schoenholz, S.S., Pennington, J., and Sohl-Dickstein, J. (2017). Deep neural networks as gaussian processes. arXiv.

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Vigilante achieved an average accuracy of 92.6% for three ambulatory behaviour identification applications.

    “For three ambulatory behaviour identification applications, they achieved an average accuracy of 92.6% [158].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [158] Lara, D., and Labrador, M.A. (2012, January 14–17). A mobile platform for real-time human activity recognition. Proceedings of the 2012 IEEE Consumer Communications and Networking Conference (CCNC), Las Vegas, NV, USA. source

      Not checked in the free run.

  • May combine two claims. The quote only partly states this claim.

    Premisecitation

    WaistonBelt X received an F1 score of 0.82 for seven activities trailed with 17 participants, with some misclassifications occurring between sitting and standing, walking and walking up and down stairs.

    “WaistonBelt X received an F1 score of 0.82, with some misclassifications occurring between sitting and standing, walking and walking up and down stairs [162].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [162] Nakamura, Y., Matsuda, Y., Arakawa, Y., and Yasumoto, K. (2019). WaistonBelt X:A Belt-Type Wearable Device with Sensing and Intervention Toward Health Behavior Change. Sensors, 19. source

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Interpretationcitation

    If an ANN system can automatically detect standardised functional tests of human movement from long-term data, then the sensors could provide a long-term measurement of patient activity and recovery at home with minimal intervention.

    “If an ANN system can automatically detect standardised functional tests of human movement from long-term data, then the sensors could provide a long-term measurement of patient activity and recovery at home with minimal intervention [156].”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [156] Neapolitan, R.E., and Jiang, X. (2018). Neural Networks and Deep Learning, Determination Press. source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Kao assessed seven activities: Driving, biking, punching, brushing teeth, knocking, swinging and using a computer, with a total accuracy of 94.71%.

    “It assessed seven activities: Driving, biking, punching, brushing teeth, knocking, swinging and using a computer. It has a total accuracy of 94.71% [160].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [160] Tzu-Ping, K., Che-Wei, L., and Jeen-Shing, W. (2009, January 5–8). Development of a portable activity detector for daily activity recognition. Proceedings of the IEEE International Symposium on Industrial Electronics, Seoul, Korea.

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Tapia et al. generates 80.6% accuracy for 30 activities, including lifting weights, doing push-ups and rowing.

    “Tapia et al. is another HAR system that generates 80.6% accuracy for 30 activities, including lifting weights, doing push-ups and rowing [159].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [159] Tapia, E.M., Intille, S.S., Haskell, W., Larson, K., Wright, J., King, A., and Friedman, R. (2007, January 11–13). Real-time recognition of physical activities and theirintensities using wireless accelerometers and a heart monitor. Proceedings of the 2007 11th IEEE International Symposium on Wearable Computers, Boston, MA, USA. source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Naive Bayes and K-Nearest Neighbour classifiers are often used to recognise seven human body motions, including jumping, walking and running.

    “Naive Bayes (NB) and K-Nearest Neighbour (KNN) classifiers are often used to recognise seven human body motions, including jumping, walking and running [165].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [165] Kaghyan, 2012, Activity recognitionusing k-nearest neighbor algorithm on smartphone with triaxial accelerometer, Int. J. Inform. Models Anal.

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    With the advancement of deep learning technology, manual feature extraction is no longer necessary, and performance improvement in complicated human activity identification is possible.

    “With the advancement of DL technology, manual feature extraction is no longer necessary, and performance improvement in complicated human activity identification is possible [150].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [150] Xu, 2018, Movement Human Actions Recognition Based on Machine Learning, Int. J. Online Eng. (iJOE) source

      Not checked in the free run.

  • May combine two claims. The quote only partly states this claim.

    Premisecitation

    The PR algorithm indicated better prediction than the Apriori algorithm, with standard deviations around 2.6 and 3.32, respectively.

    “The PR-Algorithm and Apriori Algorithm Standard Deviations were around 2.6 and 3.32, respectively.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [168] Gupta, S.M., and Mujawar, A. (2018, January 11–12). Tracking and Prediciting Movement Patterns of a Moving Object in Wiresless Sensor Network. Proceedings of the 2018 2nd International Conference on Trends in Electronics and Informatics (ICOEI), Tirunelveli, India. source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    InnoHAR demonstrated significantly higher performance and strong generalisation performance than state-of-the-art and baselines on a sliding window dataset.

    “On a sliding window dataset, InnoHAR demonstrated significantly higher performance and strong generalisation performance than state-of-the-art and baselines.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [170] Xu, 2019, InnoHAR: A Deep Neural Network for Complex Human Activity Recognition, IEEE Access source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Integrating more body parts increased the classification accuracy of the NN system.

    “According to this study, integrating more body parts increased the classification accuracy of their NN system [171].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [171] Clouthier, 2020, Sensor Data Required for Automatic Recognition of Athletic Tasks Using Deep Neural Networks, Front. Bioeng. Biotechnol. source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    The F-Score for the Sanitation dataset was 0.998 for Run and 0.905 for a walk.

    “The F-Score for the Sanitation dataset was 0.998 for Run and 0.905 for a walk.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [169] Zhang, 2019, Human Activity Recognition Based on Motion Sensor Using U-Net, IEEE Access source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    InnoHAR was generalised and validated using three datasets: Opportunity, PAMAP2, and Smartphone, with F-scores of 0.946, 0.935 and 0.945, respectively.

    “This model has been generalised and validated using three datasets: Opportunity, PAMAP2, and Smartphone, with F-scores of 0.946, 0.935 and 0.945, respectively [170].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [170] Xu, 2019, InnoHAR: A Deep Neural Network for Complex Human Activity Recognition, IEEE Access source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Non-parametric discovery of human movement patterns for walking, sedentary activities, running and jumping from accelerometer data using a Hierarchical Dirichlet process model implemented using Support Vector Machine exhibited precision of 0.81 and recall of 0.77 for sliding window time durations.

    “Non-parametric discovery of human movement patterns for walking, sedentary activities, running and jumping from accelerometer data using a Hierarchical Dirichlet process (HDP) model implemented using Support Vector Machine (SVM) exhibited precision of 0.81 and recall of 0.77 for sliding window time durations.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [135] Nguyen, 2016, Nonparametric discovery of movement patterns from accelerometer signals, Pattern Recognit. Lett. source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Feed-forward ANN had a 15% error rate, whereas the LSTM exhibited a 9% improvement in accuracy.

    “The Feed-forward ANN had a 15% error rate, whereas the LSTM exhibited a 9% improvement in accuracy [175].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [175] Choi, A., Jung, H., and Mun, J.H. (2019). Single Inertial Sensor-Based Neural Networks to Estimate COM-COP Inclination Angle during Walking. Sensors, 19. source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    The RCE NN-based HGR system had a 98.6% accuracy rate.

    “This sort of technology had a 98.6% accuracy rate [182].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [182] Kim, M., Cho, J., Lee, S., and Jung, Y. (2019). IMU Sensor-Based Hand Gesture Recognition for Human-Machine Interfaces. Sensors, 19. source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Multivariate analyses show that poor sleep scores cause more medical expenses and depressive symptoms.

    “Multivariate analyses show that poor sleep scores cause more medical expenses and depressive symptoms [191].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [191] Duffy, 2013, HHS Public Access Author manuscript, J. Community Health

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Completely trained models for surface and age with complete IMU data had good precision (96.4, 95.2%), recall (96.3, 94.7%), f1-score (96.3, 94.6%) and accuracy (96.3, 94.6%).

    “The results revealed that completely trained models for surface and age with complete IMU data had good precision (96.4, 95.2%), recall (96.3, 94.7%), f1-score (96.3, 94.6%) and accuracy (96.3, 94.6%). (96.3, 94.7%) [186].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [186] Hu, 2018, Machine learning algorithms based on signals from a single wearable inertial sensor can detect surface- and age-related differences in walking, J. Biomech. source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S4 (based).

    Interpretationnone given

    Data collected during an occasional clinical assessment represents a very limited snapshot of a patient's physiological condition.

    “Data collected during an occasional clinical assessment represents a very limited snapshot of a patient’s physiological condition.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Data gloves with integrated IMU sensors reached a speed of 8.94 milliseconds per frame and an accuracy of 98% with 0.1 s of predicting.

    “With 0.1 s of predicting, the system reached a speed of 8.94 milliseconds per frame and an accuracy of 98% [183].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [183] DiLiberti, N., Peng, C., Kaufman, C., Dong, Y., and Hansberger, J.T. (2019, January 21–25). Real-Time Gesture Recognition Using 3D Sensory Data and a Light Convolutional Neural Network. Proceedings of the 27th ACM International Conference on Multimedia, Nice, France. source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S4 (based).

    Interpretationnone given

    Wearable devices can enable users to assess their own health.

    “Wearable devices can enable users to assess their own health.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Only a small body of evidence exists to indicate that raw accelerometer data are an accurate representation of body acceleration and most scientific evidence indicates that the validity of accelerometer data are dependent mostly on epoch averages.

    “Unfortunately, only a small body of evidence exists to indicate that raw accelerometer data are an accurate representation of body acceleration and most scientific evidence indicates that the validity of accelerometer data are dependent mostly on epoch averages [192].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [192] Migueles, 2019, GGIR: A Research Community–Driven Open Source R Package for Generating Physical Activity and Sleep Outcomes from Multi-Day Raw Accelerometer Data, J. Meas. Phys. Behav. source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

    Interpretationnone given

    If all human activities can be extracted automatically from ambulatory gestures, patient monitoring effort can be considerably reduced.

    “If all human activities can be extracted automatically from ambulatory gestures, patient monitoring effort can be considerably reduced.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

    Interpretationnone given

    Analysing and identifying specific types of activity within sensor data can further enhance the assessment of the patient's quality of daily life.

    “Analysing and identifying specific types of activity within sensor data can further enhance the assessment of the patient’s quality of daily life.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

    Interpretationreasoning

    Developing an efficient automated activity detection system is a difficult task.

    “As a result, developing an efficient automated activity detection system is a difficult task.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: The time required to accomplish each activity varies by gender, age and health condition.

  • Probably not a substantial claim: may not pass S4 (based).

    Interpretationnone given

    These data are rich in objective detail when compared to clinical assessment data.

    “These data are rich in objective detail when compared to clinical assessment data.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Another research study for in-home polysomnography using the Cole-Kripke algorithm on a sample of 35 healthy volunteers found that it was more sensitive to identify sleep (88-96%), but less specific to detect awake time (35-64%).

    “Another research study for in-home polysomnography (PSG) using the Cole–Kripke algorithm on a sample of 35 healthy volunteers found that it was more sensitive to identify sleep (88–96%), but less specific to detect awake time (35–64%) [195].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [195] Quante, 2018, Actigraphy-based sleep estimation in adolescents and adults: A comparison with polysomnography using two scoring algorithms, Nat. Sci. Sleep source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S4 (based).

    Interpretationnone given

    Wearable sensors can passively collect information about a patient's daily activities.

    “Wearable sensors can passively collect information about a patient’s daily activities, and hence.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    The Cole-Kripke algorithm undervalued sleep and exaggerated wakefulness in children and vice versa in adults.

    “Previous research indicates that this algorithm undervalued sleep and exaggerated wakefulness in children and vice versa in adults.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [195] Quante, 2018, Actigraphy-based sleep estimation in adolescents and adults: A comparison with polysomnography using two scoring algorithms, Nat. Sci. Sleep source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

    Interpretationnone given

    Both GGIR and Cole-Kripke algorithms do not provide facilities to segment data based on specific movement patterns.

    “Both GGIR and Cole–Kripke algorithms do not provide facilities to segment data based on specific movement patterns.”
    Quote in the paperNot a substantial claim
68 of 68

  • The paper relies on this without support.

    Long-term assessment of ambulatory movement with wearable sensors produces very large data volumes.

    “The volume of data collected by wearable sensors during long-term assessment of ambulatory movement can become immense in tuple size.”

  • The paper relies on this without support.

    Wearable sensors can typically assess and quantify the wearer’s physiology.

    “Wearable sensors can typically assess and quantify the wearer’s physiology and are commonly employed for human activity detection and quantified self-assessment.”

  • The paper relies on this without support.

    Wearable sensors are increasingly utilised to monitor patient health, rapidly assist with disease diagnosis, and help predict and often improve patient outcomes.

    “Wearable sensors are increasingly utilised to monitor patient health, rapidly assist with disease diagnosis, and help predict and often improve patient outcomes.”

  • The paper relies on this without support.

    Many wearable devices store data in local storage without encryption or data protection.

    “Many wearable devices store data in local storage without encryption or data protection.”

  • The paper relies on this without support.

    Wearable sensors are always synchronised with smartphones for data transfer.

    “Wearable sensors are always synchronised with smartphones for data transfer, and third-party apps installed on smartphones can increase the susceptibility to data hacking.”

  • The paper relies on this without support.

    Third-party apps installed on smartphones can increase the susceptibility to data hacking.

    “Wearable sensors are always synchronised with smartphones for data transfer, and third-party apps installed on smartphones can increase the susceptibility to data hacking.”

  • The paper relies on this without support.

    In a passive attack, a possible intruder can easily obtain data without the users’ knowledge, due to a lack of security on the wearable device communication pathway.

    “In a passive attack, a possible intruder can easily obtain data without the users’ knowledge, due to a lack of security on the wearable device communication pathway.”

  • The paper relies on this without support.

    Many fitness trackers save the user’s running, walking, or cycling routes by tracking GPS coordinates.

    “Many fitness trackers save the user’s running, walking, or cycling routes by tracking GPS coordinates.”

  • The paper relies on this without support.

    Microphones can trigger a privacy risk by capturing unauthorised audio or recording the voices of others without their consent.

    “Microphones can trigger a privacy risk by capturing unauthorised audio or recording the voices of others without their consent.”

  • The paper relies on this without support.

    A wearable camera could also be hacked, exposing the user’s personal information and current surroundings.

    “A wearable camera could also be hacked, exposing the user’s personal information and current surroundings.”

  • The paper relies on this without support.

    Health information is sensitive or confidential in nature.

    “Since health information is sensitive or confidential in nature, data privacy and security are critical in healthcare applications that use wearable devices.”

  • The paper relies on this without support.

    Supervised learning is mainly applicable for the classification of a discrete-class problem.

    “Supervised learning is mainly applicable for the classification of a discrete-class problem.”

  • The paper relies on this without support.

    Human motion characteristics vary from person to person.

    “Human motion characteristics vary from person to person.”

  • The paper relies on this without support.

    Each person’s motions of body components, such as the arm, leg and head for regular activities, such as walking, jogging and eating, are unique.

    “Each person’s motions of body components, such as the arm, leg and head for regular activities, such as walking, jogging and eating, are unique.”

  • The paper relies on this without support.

    The time required to do each of these activities also varies between individuals.

    “The time required to do each of these activities also varies between individuals.”

  • The paper relies on this without support.

    Decisions about the patient’s health condition, status of disease and recommended types of treatments and medical procedures are concluded by comparing data recorded during a clinical assessment to general population averages.

    “Decisions about the patient’s health condition, status of disease and recommended types of treatments and medical procedures are concluded by comparing data recorded during a clinical assessment to general population averages.”

  • The paper relies on this without support.

    The time required to accomplish each activity varies by gender, age and health condition.

    “The time required to accomplish each activity varies by gender, age and health condition.”
17 of 17

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Head-gesture classifier has two reported accuracies

    Minor

    For the same VRH study using 975 head-movement patterns from 12 people, the text reports 91% accuracy, while Table 6 reports a classification rate of 0.975 (97.5%).

1 of 1

Cites

Flagged first, then by citations

Research works printed but not linked: 14

  1. Frost, 2014, Market Research Report, Strategy R, 7215, 1
  2. Kumar, 2019, Wearable Technology in Combination with Diabetes, Int. J. Res. Eng. Sci. Manag., 2, 1
  3. 2015, Wearable Technologies and Its Future Applications, Int. J. Electr. Electron. Data Commun., 3, 45
  4. (2021, June 01). Fitbit Sense In-Depth Review_All the Data Without the Clarity_DC Rainmaker. Available online: www.dcrainmaker.com/2020/09/fitbit-sense-in-depth-review-all-the-data-without-the-clarity.html.
  5. (2021, June 01). Apps and Fitness–Sony Smartwatch 3 Review TechRadar. Available online: www.techradar.com/reviews/sony-smartwatch-3/4.
  6. Wilson, S., and Laing, R.M. (2018, January 23–26). Wearable Technology: Present and Future. Proceedings of the 91st World Conference, Leeds, UK.
  7. Tim, 2018, Getting the Most out of Wearable Technology in Clinical Research, J. Clin. Stud., 10, 54
  8. (2021, May 12). How Accurate Can RFID Tracking Be RFID Journal. Available online: www.rfidjournal.com/question/how-accurate-can-rfid-tracking-be.
  9. Mendu, S., and Baee, S. (2020). Redesigning the Quantified Self Ecosystem with Mental Health in Mind, ACM.
  10. Sarkania, 2013, Android Internals, Int. J. Adv. Res. Comput. Sci. Softw. Eng., 3, 143
  11. Gmyzin, D. (2017). A Comparison of Supervised Machine Learning Classification Techniques and Theory-Driven Approaches for the Prediction of Subjective Mental Workload Subjective Mental. [Master’s Thesis, Technological University Dublin].
  12. Ben-gal, I. (2014). Outlier detection Irad Ben-Gal Department of Industrial Engineering. Data Mining and Knowledge Discovery Handbook, Springer.
  13. Kaghyan, 2012, Activity recognitionusing k-nearest neighbor algorithm on smartphone with triaxial accelerometer, Int. J. Inform. Models Anal., 1, 146
  14. Duffy, 2013, HHS Public Access Author manuscript, J. Community Health, 38, 597

Cited by

Sorted by citations

Related

Authors
Vijayan, V., Connolly, J. P., Condell, J., McKelvey, N., & Gardiner, P.
Affiliations
  • Letterkenny Institute of Technology (IE)
  • Ulster University, Derry-Londonderry Campus (GB)
  • Altnagelvin Area Hospital (GB)
Corresponding author
  • Vijayan, Vini
  • Connolly, James P.
ORCID
Venue
Sensors
Publisher
MDPI AG
Published
2021-08-19
Volume
21
Issue
16
Pages
5589
ISSN
1424-8220 (ISSN-L)
DOI
10.3390/s21165589
PMC
PMC8402237
OpenAlex
W3194077052
Semantic Scholar
97ac44565da06016cec550bcff3483f011c09fd6
Cited by
435
References
148
Citations by year
2026: 82 · 2025: 117 · 2024: 116 · 2023: 80 · 2022: 38 · 2021: 2
FWCI
3.68
Type
journal article
Language
en
Open access
Licence
Locations
  • Sensors (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • Ulster University Research Portal (Ulster University) (publishedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Topics
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Context-Aware Activity Recognition Systems
  • Physical Sciences › Engineering › Biomedical Engineering › Non-Invasive Vital Sign Monitoring
  • Physical Sciences › Computer Science › Computer Networks and Communications › Wireless Sensor Networks for Data Analysis
Keywords
wearable sensors, connected health, human activity detection, motion tracking, sleep monitoring, activity monitoring, long-term data collection, patient monitoring, functional ability assessment, quantified self
MeSH terms
Wearable Electronic Devices, Humans, Monitoring, Physiologic, Movement, Sleep, Self Report

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-09.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-09.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review