Validity of the Apple Watch® for monitoring push counts in people using manual wheelchairs

Emerging methods for measuring physical activity using accelerometry in children and adolescents with neuromotor disorders: a narrative review

BACKGROUND

Children and adolescents with neuromotor disorders need regular physical activity to maintain optimal health and functional independence throughout their development. To this end, reliable measures of physical activity are integral to both assessing habitual physical activity and testing the efficacy of the many interventions designed to increase physical activity in these children. Wearable accelerometers have been used for children with neuromotor disorders for decades; however, studies most often use disorder-specific cut points to categorize physical activity intensity, which lack generalizability to a free-living environment. No reviews of accelerometer data processing methods have discussed the novel use of machine learning techniques for monitoring physical activity in children with neuromotor disorders.

METHODS

In this narrative review, we discuss traditional measures of physical activity (including questionnaires and objective accelerometry measures), the limitations of standard analysis for accelerometry in this unique population, and the potential benefits of applying machine learning approaches. We also provide recommendations for using machine learning approaches to monitor physical activity.

CONCLUSIONS

While wearable accelerometers provided a much-needed method to quantify physical activity, standard cut point analyses have limitations in children with neuromotor disorders. Machine learning models are a more robust method of analyzing accelerometer data in pediatric neuromotor disorders and using these methods over disorder-specific cut points is likely to improve accuracy of classifying both type and intensity of physical activity. Notably, there remains a critical need for further development of classifiers for children with more severe motor impairments, preschool aged children, and children in hospital settings.

Development of SkinTracker, an integrated dermatology mobile app and web portal enabling remote clinical research studies

Introduction

In-person dermatology clinical research studies often face recruitment and participation challenges due to travel-, time-, and cost-associated barriers. Studies incorporating virtual/asynchronous formats can potentially enhance research subject participation and satisfaction, but few mobile health tools are available to enable remote study conduct. We developed SkinTracker, a patient-facing mobile app and researcher-facing web platform, that enables longitudinal collection of skin photos, patient reported outcomes, and biometric health and environmental data.

Methods

Eight design thinking sessions including dermatologists, clinical research staff, software engineers, and graphic designers were held to create the components of SkinTracker. Following iterative prototyping, SkinTracker was piloted across six adult and four pediatric subjects with atopic dermatitis (AD) of varying severity levels to test and provide feedback on SkinTracker for six months.

Results

The SkinTracker app enables collection of informed consent for study participation, baseline medical history, standardized skin photographs, patient-reported outcomes (e.g., Patient Oriented Eczema Measure (POEM), Pruritus Numerical Rating Scale (NRS), Dermatology Life Quality Index (DLQI)), medication use, adverse events, voice diary to document qualitative experiences, chat function for communication with research team, environmental and biometric data such as exercise and sleep metrics through integration with an Apple Watch. The researcher web portal allows for management and visualization of subject enrollment, skin photographs for examination and severity scoring, survey completion, and other patient modules. The pilot study requested that subjects complete surveys and photographs on a weekly to monthly basis via the SkinTracker app. Afterwards, participants rated their experience in a 7-item user experience survey covering app function, design, and desire for participation in future studies using SkinTracker. Almost all subjects agreed or strongly agreed that SkinTracker enabled more convenient participation in skin research studies compared to an in-person format.

Discussion

To our knowledge, SkinTracker is one of the first integrated app- and web-based platforms allowing collection and management of data commonly obtained in clinical research studies. SkinTracker enables detailed, frequent capture of data that may better reflect the fluctuating course of conditions such as AD, and can be modularly customized for different skin conditions to improve dermatologic research participation and patient access.

Investigating the Accuracy of Wheelchair Push Counts Measured by Fitness Watches: A Systematic Review

Wheelchair users face an elevated risk of metabolic syndromes due to their sedentary lifestyles. One of the methods to prevent and treat various metabolic syndromes is regular physical activity, which varies among individuals based on their abilities. Monitoring physical activity among them can be performed by using wearable physical activity monitors (WPAMs), which utilize accelerometers and algorithms to track wheelchair push counts. However, the accuracy of push count detection varies among the devices due to technological limitations. The objective of this literature review was to evaluate the accuracy of WPAMs, specifically smartwatches, in measuring physical activity in the wheelchair population. This systematic literature review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. The databases PubMed, Embase, and Cumulative Index to Nursing and Allied Health Literature (CINAHL) were searched in November 2022 for relevant articles. The initial search yielded 447 articles, seven of which were selected based on the inclusion criteria, which were as follows: participant ability to maneuver a wheelchair, arm- or wrist-worn WPAMs, and articles published after 2017. Among the devices studied, the Apple Watch was determined to be the most accurate calibration system for wheelchair users, with the lowest mean absolute percentage error (MAPE). Each succeeding generation of the Apple Watch (first to fourth) studied was more accurate than the previous. The review demonstrates that research on wheelchair fitness tracking remains scarce and further studies are required to address this issue.

Real-Life Wheelchair Mobility Metrics from IMUs

Daily wheelchair ambulation is seen as a risk factor for shoulder problems, which are prevalent in manual wheelchair users. To examine the long-term effect of shoulder load from daily wheelchair ambulation on shoulder problems, quantification is required in real-life settings. In this study, we describe and validate a comprehensive and unobtrusive methodology to derive clinically relevant wheelchair mobility metrics (WCMMs) from inertial measurement systems (IMUs) placed on the wheelchair frame and wheel in real-life settings. The set of WCMMs includes distance covered by the wheelchair, linear velocity of the wheelchair, number and duration of pushes, number and magnitude of turns and inclination of the wheelchair when on a slope. Data are collected from ten able-bodied participants, trained in wheelchair-related activities, who followed a 40 min course over the campus. The IMU-derived WCMMs are validated against accepted reference methods such as Smartwheel and video analysis. Intraclass correlation (ICC) is applied to test the reliability of the IMU method. IMU-derived push duration appeared to be less comparable with Smartwheel estimates, as it measures the effect of all energy applied to the wheelchair (including thorax and upper extremity movements), whereas the Smartwheel only measures forces and torques applied by the hand at the rim. All other WCMMs can be reliably estimated from real-life IMU data, with small errors and high ICCs, which opens the way to further examine real-life behavior in wheelchair ambulation with respect to shoulder loading. Moreover, WCMMs can be applied to other applications, including health tracking for individual interest or in therapy settings.

Accuracy and Precision of Actigraphy and SMARTwheels for Measuring Push Counts Across a Series of Wheelchair Propulsion Trials in Non-disabled Young Adults

Background: There has been a growing interest in “Lifestyle Physical Activity” (LPA) among wheelchair users. LPA can be quantified via “pushes” as an outcome metric. This study examined the accuracy and precision of research-grade devices for counting pushes across a series of wheelchair propulsion trials. Methods: Eleven non-disabled, young adults completed 19, 1-minute wheelchair propulsion trials at self-selected speeds with a wheelchair equipped with a SMARTwheel (SW) device while being video recorded. Participants also wore 2 ActiGraph accelerometers, one on the wrist and one on the upper arm. Video footage enabled manual counting of the number of pushes (gold standard). Total pushes were averaged across 16 workloads (3 trials of repeated workloads were excluded) for each device and compared to manually counted pushes. Results: Compared to manually counted pushes, SW demonstrated the greatest accuracy (mean difference [MD] compared to video of 2.3 pushes [4.5% error]) and precision (standard deviation of the mean difference [SDMD]) compared to video of 4 pushes, (Coefficient of Variation [CV] =.04), followed by the upper arm-worn accelerometer (MD of 4.4 pushes [10.4% error] and SDMD of 10, [CV= .06]) and the wrist-worn accelerometer (MD of 12.6 pushes [27.8% error] and SDMD of 13 [CV=.15]). Conclusions: SW demonstrated greater accuracy and precision than ActiGraph accelerometers placed on the upper arm and wrist. The accelerometer placed on the upper arm was more accurate and precise than the accelerometer placed on the wrist. Future investigations should be conducted to identify the source(s) of inaccuracy among wearable push counters.

Measurement Performance of Activity Measurements with Newer Generation of Apple Watch in Wheelchair Users with Spinal Cord Injury

Background  The level of physical activity (PA) of people with spinal cord injury (SCI) has an impact on long-term complications. Currently, PA is mostly assessed by interviews. Wearable activity trackers are promising tools to objectively measure PA under everyday conditions. The only off-the-shelf, wearable activity tracker with specific measures for wheelchair users is the Apple Watch.

Objectives  This study analyzes the measurement performance of Apple Watch Series 4 for wheelchair users and compares it with an earlier generation of the device.

Methods  Fifteen participants with subacute SCI during their first in-patient phase followed a test course using their wheelchair. The number of wheelchair pushes was counted manually by visual inspection and with the Apple Watch. Difference between the Apple Watch and the rater was analyzed with mean absolute percent error (MAPE) and a Bland–Altman plot. To compare the measurement error of Series 4 and an older generation of the device a t -test was calculated using data for Series 1 from a former study.

Results  The average of differences was 12.33 pushes ( n  = 15), whereas participants pushed the wheelchair 138.4 times on average (range 86–271 pushes). The range of difference and the Bland–Altman plot indicate an overestimation by Apple Watch. MAPE is 9.20% and the t -test, testing for an effect of Series 4 on the percentage of error compared with Series 1, was significant with p  < 0.05.

Conclusion  Series 4 shows a significant improvement in measurement performance compared with Series 1. Series 4 can be considered as a promising data source to capture the number of wheelchair pushes on even grounds. Future research should analyze the long-term measurement performance during everyday conditions of Series 4.

Detecting clinical practice guideline-recommended wheelchair propulsion patterns with wearable devices following a wheelchair propulsion intervention

Wheelchair propulsion interventions typically teach manual wheelchair users to perform wheelchair propulsion biomechanics as recommended by the Clinical Practice Guidelines (CPG). Outcome measures for these interventions are primarily laboratory based. Discrepancies remain between manual wheelchair propulsion (MWP) in laboratory-based examinations and propulsion in the real-world. Current developments in machine learning (ML) allow for monitoring of MWP in the real world. In this study, we collected data from participants enrolled in two wheelchair propulsion interventions, then built an ML algorithm to distinguish CPG recommended MWP patterns from non-CPG-recommended patterns. Eight primary manual wheelchair users did not initially follow CPG recommendations but learned and performed CPG propulsion after the interventions. Participants each wore two inertial measurement units as they propelled their wheelchairs on a roller system, indoors overground, and outdoors. ML models were trained to classify propulsion patterns as following the CPG or not following the CPG. Video recordings were used for reference. For indoor detection, we found that a subject-independent model was able to achieve 85% accuracy. For outdoor detection, we found that the subject-independent model achieved 75.4% accuracy. These results provide further evidence that CPG and non-CPG-recommended MWP patterns can be predicted with wearable sensors using an ML algorithm.

Lessons from Use of Continuous Glucose Monitoring Systems in Digital Healthcare

We live in a digital world where a variety of wearable medical devices are available. These technologies enable us to measure our health in our daily lives. It is increasingly possible to manage our own health directly through data gathered from these wearable devices. Likewise, healthcare professionals have also been able to indirectly monitor patients' health. Healthcare professionals have accepted that digital technologies will play an increasingly important role in healthcare. Wearable technologies allow better collection of personal medical data, which healthcare professionals can use to improve the quality of healthcare provided to the public. The use of continuous glucose monitoring systems (CGMS) is the most representative and desirable case in the adoption of digital technology in healthcare. Using the case of CGMS and examining its use from the perspective of healthcare professionals, this paper discusses the necessary adjustments required in clinical practices. There is a need for various stakeholders, such as medical staff, patients, industry partners, and policy-makers, to utilize and harness the potential of digital technology.