An open-source tool to identify active travel from hip-worn accelerometer, GPS and GIS data

Understanding the relationships between 24-hour movement behavior, community mobility and the neighborhood built environment for healthy aging in Brazil: The EpiMove study protocol

BACKGROUND

The population is aging rapidly worldwide, impacting public health, with countries in the Global South, such as Brazil, aging faster than developed nations. The 24-hour movement behavior is crucial for healthy aging, but its relationship with the neighborhood built environment is underresearched, especially for older adults. The EpiMove Study uses accelerometers and GPS to investigate the relationships between 24-hour movement behavior, community mobility and the neighborhood built environment for healthy aging in older Brazilian adults.

METHODS

The EpiMove Study is a representative cross-sectional study of older adults aged 60 years and older from an urban area in the southern region of Brazil. It consists of two phases. Phase 1 involves conducting home interviews to gather subjective measures of the neighborhood built environment and physical activity. Phase 2 involves delivering devices to participants' homes and collecting objective data on 24-hour movement behavior via wrist-worn wGT3X-BT ActiGraph accelerometers and community-based active transportation via hip-mounted GPS Qstarz-1000XT devices. The data are collected simultaneously over seven consecutive days, along with the participants' reasons for adhering to the study protocol.

DISCUSSION

The EpiMove study will provide a better understanding of the relationships between the perceived neighborhood environment and 24-hour movement behaviors and community-based active transportation among older adults, with a particular focus on whether environmental factors influence these behaviors, which are crucial for healthy aging. The results from the EpiMove study could offer essential evidence for developing public policies and urban interventions that support healthier and more equitable environments for aging populations, particularly in rapidly urbanizing regions.

Raw Photoplethysmography as an Enhancement for Research-Grade Wearable Activity Monitors

Wearable monitors continue to play a critical role in scientific assessments of physical activity. Recently, research-grade monitors have begun providing raw data from photoplethysmography (PPG) alongside standard raw data from inertial sensors (accelerometers and gyroscopes). Raw PPG enables granular and transparent estimation of cardiovascular parameters such as heart rate, thus presenting a valuable alternative to standard PPG methodologies (most of which rely on consumer-grade monitors that provide only coarse output from proprietary algorithms). The implications for physical activity assessment are tremendous, since it is now feasible to monitor granular and concurrent trends in both movement and cardiovascular physiology using a single noninvasive device. However, new users must also be aware of challenges and limitations that accompany the use of raw PPG data. This viewpoint paper therefore orients new users to the opportunities and challenges of raw PPG data by presenting its mechanics, pitfalls, and availability, as well as its parallels and synergies with inertial sensors. This includes discussion of specific applications to the prediction of energy expenditure, activity type, and 24-hour movement behaviors, with an emphasis on areas in which raw PPG data may help resolve known issues with inertial sensing (eg, measurement during cycling activities). We also discuss how the impact of raw PPG data can be maximized through the use of open-source tools when developing and disseminating new methods, similar to current standards for raw accelerometer and gyroscope data. Collectively, our comments show the strong potential of raw PPG data to enhance the use of research-grade wearable activity monitors in science over the coming years.

Classification of human walking context using a single-point accelerometer

Real-world walking data offers rich insights into a person's mobility. Yet, daily life variations can alter these patterns, making the data challenging to interpret. As such, it is essential to integrate context for the extraction of meaningful information from real-world movement data. In this work, we leveraged the relationship between the characteristics of a walking bout and context to build a classification algorithm to distinguish between indoor and outdoor walks. We used data from 20 participants wearing an accelerometer on the thigh over a week. Their walking bouts were isolated and labeled using GPS and self-reporting data. We trained and validated two machine learning models, random forest and ensemble Support Vector Machine, using a leave-one-participant-out validation scheme on 15 subjects. The 5 remaining subjects were used as a testing set to choose a final model. The chosen model achieved an accuracy of 0.941, an F1-score of 0.963, and an AUROC of 0.931. This validated model was then used to label the walks from a different dataset with 15 participants wearing the same accelerometer. Finally, we characterized the differences between indoor and outdoor walks using the ensemble of the data. We found that participants walked significantly faster, longer, and more continuously when walking outdoors compared to indoors. These results demonstrate how movement data alone can be used to obtain accurate information on important contextual factors. These factors can then be leveraged to enhance our understanding and interpretation of real-world movement data, providing deeper insights into a person's health.

Systematic review of best practices for GPS data usage, processing, and linkage in health, exposure science and environmental context research

Global Positioning System (GPS) technology is increasingly used in health research to capture individual mobility and contextual and environmental exposures. However, the tools, techniques and decisions for using GPS data vary from study to study, making comparisons and reproducibility challenging.

OBJECTIVES

The objectives of this systematic review were to (1) identify best practices for GPS data collection and processing; (2) quantify reporting of best practices in published studies; and (3) discuss examples found in reviewed manuscripts that future researchers may employ for reporting GPS data usage, processing and linkage of GPS data in health studies.

DESIGN

A systematic review.

DATA SOURCES

Electronic databases searched (24 October 2023) were PubMed, Scopus and Web of Science (PROSPERO ID: CRD42022322166).

ELIGIBILITY CRITERIA

Included peer-reviewed studies published in English met at least one of the criteria: (1) protocols involving GPS for exposure/context and human health research purposes and containing empirical data; (2) linkage of GPS data to other data intended for research on contextual influences on health; (3) associations between GPS-measured mobility or exposures and health; (4) derived variable methods using GPS data in health research; or (5) comparison of GPS tracking with other methods (eg, travel diary).

DATA EXTRACTION AND SYNTHESIS

We examined 157 manuscripts for reporting of best practices including wear time, sampling frequency, data validity, noise/signal loss and data linkage to assess risk of bias.

RESULTS

We found that 6% of the studies did not disclose the GPS device model used, only 12.1% reported the per cent of GPS data lost by signal loss, only 15.7% reported the per cent of GPS data considered to be noise and only 68.2% reported the inclusion criteria for their data.

CONCLUSIONS

Our recommendations for reporting on GPS usage, processing and linkage may be transferrable to other geospatial devices, with the hope of promoting transparency and reproducibility in this research.

PROSPERO REGISTRATION NUMBER

CRD42022322166.

Using Lower Limb Wearable Sensors to Identify Gait Modalities: A Machine-Learning-Based Approach

Real-world gait analysis can aid in clinical assessments and influence related interventions, free from the restrictions of a laboratory setting. Using individual accelerometers, we aimed to use a simple machine learning method to quantify the performance of the discrimination between three self-selected cyclical locomotion types using accelerometers placed at frequently referenced attachment locations. Thirty-five participants walked along a 10 m walkway at three different speeds. Triaxial accelerometers were attached to the sacrum, thighs and shanks. Slabs of magnitude, three-second-long accelerometer data were transformed into two-dimensional Fourier spectra. Principal component analysis was undertaken for data reduction and feature selection, followed by discriminant function analysis for classification. Accuracy was quantified by calculating scalar accounting for the distances between the three centroids and the scatter of each category's cloud. The algorithm could successfully discriminate between gait modalities with 91% accuracy at the sacrum, 90% at the shanks and 87% at the thighs. Modalities were discriminated with high accuracy in all three sensor locations, where the most accurate location was the sacrum. Future research will focus on optimising the data processing of information from sensor locations that are advantageous for practical reasons, e.g., shank for prosthetic and orthotic devices.

Criterion validity of wrist accelerometry for assessing energy intake via the intake-balance technique

BACKGROUND

Intake-balance assessments measure energy intake (EI) by summing energy expenditure (EE) with concurrent change in energy storage (ΔES). Prior work has not examined the validity of such calculations when EE is estimated via open-source techniques for research-grade accelerometry devices. The purpose of this study was to test the criterion validity of accelerometry-based intake-balance methods for a wrist-worn ActiGraph device.

METHODS

Healthy adults (n = 24) completed two 14-day measurement periods while wearing an ActiGraph accelerometer on the non-dominant wrist. During each period, criterion values of EI were determined based on ΔES measured by dual X-ray absorptiometry and EE measured by doubly labeled water. A total of 11 prediction methods were tested, 8 derived from the accelerometer and 3 from non-accelerometry methods (e.g., diet recall; included for comparison). Group-level validity was assessed through mean bias, while individual-level validity was assessed through mean absolute error, mean absolute percentage error, and Bland-Altman analysis.

RESULTS

Mean bias for the three best accelerometry-based methods ranged from -167 to 124 kcal/day, versus -104 to 134 kcal/day for the non-accelerometry-based methods. The same three accelerometry-based methods had mean absolute error of 323-362 kcal/day and mean absolute percentage error of 18.1-19.3%, versus 353-464 kcal/day and 19.5-24.4% for the non-accelerometry-based methods. All 11 methods demonstrated systematic bias in the Bland-Altman analysis.

CONCLUSIONS

Accelerometry-based intake-balance methods have promise for advancing EI assessment, but ongoing refinement is necessary. We provide an R package to facilitate implementation and refinement of accelerometry-based methods in future research (see paulhibbing.com/IntakeBalance).

A Scoping Review of the Use of Machine Learning in Health Economics and Outcomes Research: Part 1-Data From Wearable Devices

OBJECTIVES

With the emerging use of machine learning (ML) techniques, there has been particular interest in using wearable data for health economics and outcomes research (HEOR). We aimed to understand the emerging patterns of how ML has been applied to wearable data in HEOR.

METHODS

We identified studies published in PubMed between January 2016 and March 2021. Studies that included at least 1 HEOR-related Medical Subject Headings term, applied an ML, and used wearable data were eligible for inclusion. Two reviewers abstracted information including ML application types and data on which ML was applied and analyzed them using descriptive analyses.

RESULTS

A total of 148 studies were identified from PubMed, among which 32 studies met the inclusion criteria. There has been an increase over time in the number of ML studies using wearable data. ML has been more frequently used for monitoring events in real time (78%) than to predict future events (22%). There has been a wide range of outcomes examined, ranging from general physical or mental health (24%) to more disease-specific outcomes (eg, disease incidence [19%] and progression [13%]) and treatment-related outcomes (eg, treatment adherence [9%] and outcomes [9%]). Data for ML models were more often derived from wearable devices with specific medical purposes (60%) than those without (40%).

CONCLUSION

There has been a wide range of applications of ML to wearable data. Both medical and nonmedical wearable devices have been used as a data source, showing the potential for providing rich data for ML studies in HEOR.

Deep Learning for Classifying Physical Activities from Accelerometer Data

Physical inactivity increases the risk of many adverse health conditions, including the world’s major non-communicable diseases, such as coronary heart disease, type 2 diabetes, and breast and colon cancers, shortening life expectancy. There are minimal medical care and personal trainers’ methods to monitor a patient’s actual physical activity types. To improve activity monitoring, we propose an artificial-intelligence-based approach to classify physical movement activity patterns. In more detail, we employ two deep learning (DL) methods, namely a deep feed-forward neural network (DNN) and a deep recurrent neural network (RNN) for this purpose. We evaluate the two models on two physical movement datasets collected from several volunteers who carried tri-axial accelerometer sensors. The first dataset is from the UCI machine learning repository, which contains 14 different activities-of-daily-life (ADL) and is collected from 16 volunteers who carried a single wrist-worn tri-axial accelerometer. The second dataset includes ten other ADLs and is gathered from eight volunteers who placed the sensors on their hips. Our experiment results show that the RNN model provides accurate performance compared to the state-of-the-art methods in classifying the fundamental movement patterns with an overall accuracy of 84.89% and an overall F1-score of 82.56%. The results indicate that our method provides the medical doctors and trainers a promising way to track and understand a patient’s physical activities precisely for better treatment.

Device-Measured and Self-Reported Active Travel Associations with Cardiovascular Disease Risk Factors in an Ethnically Diverse Sample of Adults

Active travel (AT) provides an opportunity to alleviate the physical inactivity and climate crises contributing to the global chronic disease burden, including cardiovascular diseases (CVD). Though AT shows promising links to reduced CVD risk, prior studies relied on self-reported AT assessment. In the present study, device-measured and self-reported AT were compared across population subgroups and relationships with CVD risk biomarkers were evaluated for both measures. The study recruited an ethnically diverse sample (N = 602, mean age 59 years, 42% Hispanic/Latino ethnicity) from neighborhoods that varied by walkability and food access. AT was assessed using concurrently collected accelerometer and GPS data and self-report data from a validated survey. Relationships with body mass index (BMI), triglycerides, high-density lipoprotein (HDL) cholesterol, blood pressure (BP), and moderate-to-vigorous physical activity (MVPA) were modeled using multivariable linear regression. Devices captured more AT than did self-report. We found differences in AT measures by population subgroups, including race, ethnicity, education, income, vehicle access, and walkability. Men had more accelerometer-measured MVPA, though women self-reported more daily minutes. Both device and survey AT measures were positively associated with total accelerometer-measured MVPA, though the relationship was stronger with device-measured AT. Device-measured AT was associated with lower BMI. No other CVD risk biomarker was associated with either AT measure. No effect modification by Hispanic/Latino ethnicity was detected. Further studies with device-based measures are warranted to better understand the relationship between AT and cardiovascular health.

Active design of built environments for increasing levels of physical activity in adults: the ENABLE London natural experiment study

Background

Low physical activity is widespread and poses a serious public health challenge both globally and in the UK. The need to increase population levels of physical activity is recognised in current health policy recommendations. There is considerable interest in whether or not the built environment influences health behaviours, particularly physical activity levels, but longitudinal evidence is limited.

Objectives

The effect of moving into East Village (the former London 2012 Olympic and Paralympic Games Athletes’ Village, repurposed on active design principles) on the levels of physical activity and adiposity, as well as other health-related and well-being outcomes among adults, was examined.

Design

The Examining Neighbourhood Activities in Built Environments in London (ENABLE London) study was a longitudinal cohort study based on a natural experiment.

Setting

East Village, London, UK.

Participants

A cohort of 1278 adults (aged ≥ 16 years) and 219 children seeking to move into social, intermediate and market-rent East Village accommodation were recruited in 2013–15 and followed up after 2 years.

Intervention

The East Village neighbourhood, the former London 2012 Olympic and Paralympic Games Athletes’ Village, is a purpose-built, mixed-use residential development specifically designed to encourage healthy active living by improving walkability and access to public transport.

Main outcome measure

Change in objectively measured daily steps from baseline to follow-up.

Methods

Change in environmental exposures associated with physical activity was assessed using Geographic Information System-derived measures. Individual objective measures of physical activity using accelerometry, body mass index and bioelectrical impedance (per cent of fat mass) were obtained, as were perceptions of change in crime and quality of the built environment. We examined changes in levels of physical activity and adiposity using multilevel models adjusting for sex, age group, ethnic group, housing sector (fixed effects) and baseline household (random effect), comparing the change in those who moved to East Village (intervention group) with the change in those who did not move to East Village (control group). Effects of housing sector (i.e. social, intermediate/affordable, market-rent) as an effect modifier were also examined. Qualitative work was carried out to provide contextual information about the perceived effects of moving to East Village.

Results

A total of 877 adults (69%) were followed up after 2 years (mean 24 months, range 19–34 months, postponed from 1 year owing to the delayed opening of East Village), of whom 50% had moved to East Village; insufficient numbers of children moved to East Village to be considered further. In adults, moving to East Village was associated with only a small, non-significant, increase in mean daily steps (154 steps, 95% confidence interval –231 to 539 steps), more so in the intermediate sector (433 steps, 95% confidence interval –175 to 1042 steps) than in the social and market-rent sectors (although differences between housing sectors were not statistically significant), despite sizeable improvements in walkability, access to public transport and neighbourhood perceptions of crime and quality of the built environment. There were no appreciable effects on time spent in moderate to vigorous physical activity or sedentary time, body mass index or percentage fat mass, either overall or by housing sector. Qualitative findings indicated that, although participants enjoyed their new homes, certain design features might actually serve to reduce levels of activity.

Conclusions

Despite strong evidence of large positive changes in neighbourhood perceptions and walkability, there was only weak evidence that moving to East Village was associated with increased physical activity. There was no evidence of an effect on markers of adiposity. Hence, improving the physical activity environment on its own may not be sufficient to increase population physical activity or other health behaviours.

Funding

This project was funded by the National Institute for Health Research (NIHR) Public Health Research programme and will be published in full in Public Health Research; Vol. 8, No. 12. See the NIHR Journals Library website for further project information. This research was also supported by project grants from the Medical Research Council National Prevention Research Initiative (MR/J000345/1).

The effect of moving to East Village, the former London 2012 Olympic and Paralympic Games Athletes' Village, on mode of travel (ENABLE London study, a natural experiment)

BACKGROUND

Interventions to encourage active modes of travel (walking, cycling) may improve physical activity levels, but longitudinal evidence is limited and major change in the built environment / travel infrastructure may be needed. East Village (the former London 2012 Olympic Games Athletes Village) has been repurposed on active design principles with improved walkability, open space and public transport and restrictions on residential car parking. We examined the effect of moving to East Village on adult travel patterns.

METHODS

One thousand two hundred seventy-eight adults (16+ years) seeking to move into social, intermediate, and market-rent East Village accommodation were recruited in 2013-2015, and followed up after 2 years. Individual objective measures of physical activity using accelerometry (ActiGraph GT3X+) and geographic location using GPS travel recorders (QStarz) were time-matched and a validated algorithm assigned four travel modes (walking, cycling, motorised vehicle, train). We examined change in time spent in different travel modes, using multilevel linear regresssion models adjusting for sex, age group, ethnicity, housing group (fixed effects) and household (random effect), comparing those who had moved to East Village at follow-up with those who did not.

RESULTS

Of 877 adults (69%) followed-up, 578 (66%) provided valid accelerometry and GPS data for at least 1 day (≥540 min) at both time points; half had moved to East Village. Despite no overall effects on physical activity levels, sizeable improvements in walkability and access to public transport in East Village resulted in decreased daily vehicle travel (8.3 mins, 95%CI 2.5,14.0), particularly in the intermediate housing group (9.6 mins, 95%CI 2.2,16.9), and increased underground travel (3.9 mins, 95%CI 1.2,6.5), more so in the market-rent group (11.5 mins, 95%CI 4.4,18.6). However, there were no effects on time spent walking or cycling.

CONCLUSION

Designing walkable neighbourhoods near high quality public transport and restrictions on car usage, may offer a community-wide strategy shift to sustainable transport modes by increasing public transport use, and reducing motor vehicle travel.

Electrically assisted cycling for individuals with type 2 diabetes mellitus: protocol for a pilot randomized controlled trial

Background

The global incidence of type 2 diabetes mellitus (T2DM) is increasing. Given the many complications associated with T2DM, effective management of the disease is crucial. Physical activity is considered to be a key component of T2DM management. However, people with T2DM are generally less physically active than individuals without T2DM and adherence to physical activity is often poor following completion of lifestyle interventions. As such, developing interventions that foster sustainable physical activity is of high priority. Electrically assisted bicycles (e-bikes) have been highlighted as a potential strategy for promoting physical activity in this population. E-bikes provide electrical assistance to the rider only when pedalling and could overcome commonly reported barriers to regular cycling. This paper describes the protocol of the PEDAL-2 pilot randomized controlled trial, an e-cycling intervention aimed at increasing physical activity in individuals with T2DM.

Methods

A parallel-group two-arm randomized waitlist-controlled pilot trial will be conducted. Forty individuals with T2DM will be randomly assigned, in a 1:1 allocation ratio, to an e-cycling intervention or waitlist control. Recruitment and screening will close once 20 participants have been randomized to each study arm. The intervention will involve e-bike training with a certified cycle instructor and provision of an e-bike for 12 weeks. Data will be collected at baseline, during the intervention and immediately post-intervention using both quantitative and qualitative methods. In this trial, the primary interests are determination of effective recruitment strategies, recruitment and consent rates, adherence and retention and delivery and receipt of the intervention. The potential impact of the intervention on a range of clinical, physiological and behaviour outcomes will be assessed to examine intervention promise. Data analyses will be descriptive.

Discussion

This paper describes the protocol for the PEDAL-2 pilot randomized controlled trial. Results from this trial will provide information on trial feasibility and identify the promise of e-cycling as a strategy to positively impact the health and behaviour of individuals with T2DM. If appropriate, this information can be used to design and deliver a fully powered definitive trial.

Trial registration

ISRCTN, ISRCTN67421464. Registered 03/01/2019.

Machine learning algorithms can classify outdoor terrain types during running using accelerometry data

BACKGROUND

Running is a popular physical activity that benefits health; however, running surface characteristics may influence loading impact and injury risk. Machine learning algorithms could automatically identify running surface from wearable motion sensors to quantify running exposures, and perhaps loading and injury risk for a runner.

RESEARCH QUESTION

(1) How accurately can machine learning algorithms identify surface type from three-dimensional accelerometer sensors? (2) Does the sensor count (single or two-sensor setup) affect model accuracy?

METHODS

Twenty-nine healthy adults (23.3 ± 3.6 years, 1.8 ± 0.1 m, and 63.6 ± 8.5 kg) participated in this study. Participants ran on three different surfaces (concrete, synthetic, woodchip) while fit with two three-dimensional accelerometers (lower-back and right tibia). Summary features (n = 208) were extracted from the accelerometer signals. Feature-based Gradient Boosting (GB) and signal-based deep learning Convolutional Neural Network (CNN) models were developed. Models were trained on 90% of the data and tested on the remaining 10%. The process was repeated five times, with data randomly shuffled between train-test splits, to quantify model performance variability.

RESULTS

All models and configurations achieved greater than 90% average accuracy. The highest performing models were the two-sensor GB and tibia-sensor CNN (average accuracy of 97.0 ± 0.7 and 96.1 ± 2.6%, respectively).

SIGNIFICANCE

Machine learning algorithms trained on running data from a single- or dual-sensor accelerometer setup can accurately distinguish between surfaces types. Automatic identification of surfaces encountered during running activities could help runners and coaches better monitor training load, improve performance, and reduce injury rates.

The future of activity space and health research

We discuss the future of activity space and health research in the context of a recently published systematic review. Our discussion outlines a number of elements for reflection among the research community. We need to think beyond activity space and reconceptualize exposure in era of high volume, high precision location data. We need to develop standardized methods for understanding global positioning system data. We must adopt replicable scientific computing processes and machine learning models. Finally, we must embrace modern notions of causality in order to contend with the conceptual challenges faced by our research field.

Socio-Ecological Natural Experiment with Randomized Controlled Trial to Promote Active Commuting to Work: Process Evaluation, Behavioral Impacts, and Changes in the Use and Quality of Walking and Cycling Paths

Active commuting to work (ACW) has beneficial effects on health, traffic, and climate. However, more robust evidence is needed on how to promote ACW. This paper reports the findings of a multilevel natural experiment with a randomized controlled trial in 16 Finnish workplaces. In Phase 1, 11 workplaces (1823 employees) from Area 1 were exposed to environmental improvements in walking and cycling paths. In Phase 2, five more workplaces (826 employees) were recruited from Area 2 and all workplaces were randomized into experimental group (EXP) promoting ACW with social and behavioral strategies and comparison group (COM) participating only in data collection. Process and impact evaluation with questionnaires, travel diaries, accelerometers, traffic calculations, and auditing were conducted. Statistics included Wilcoxon Signed Ranks Test, Mann-Whitney U-test, and after-before differences with 95% confidence intervals (95% CI). After Phase 1, positive change was seen in the self-reported number of days, which the employees intended to cycle part of their journey to work in the following week (p = 0.001). After Phase 2, intervention effect was observed in the proportion of employees, who reported willingness to increase walking (8.7%; 95% CI 1.8 to 15.6) and cycling (5.5%; 2.2 to 8.8) and opportunity to cycle part of their journey to work (5.9%; 2.1 to 9.7). To conclude, the intervention facilitated employees’ motivation for ACW, which is the first step towards behavior change.