Outdoor Walking Speeds of Apparently Healthy Adults: A Systematic Review and Meta-analysis

Ankle Kinematics Estimation Using Artificial Neural Network and Multimodal IMU Data

Inertial measurement units (IMUs) have become attractive for monitoring joint kinematics due to their portability and versatility. However, their limited accuracy, inability to analyze data in real-time, and complex data fusion algorithms requiring precise sensor-to-segment calibrations hinder their clinical and daily use. This paper introduces KEEN (KinEmatics Estimation Network), an innovative framework that exploits lightweight artificial neural networks (ANNs) to provide real-time predictions of multi-plane ankle kinematics using a minimal number of IMUs, without calibration requirements. Five ANN algorithms were developed and evaluated using 42 inputs derived from four IMUs in both intra-subject and inter-subject tasks. Extensive experimental results yielded exciting findings: even a single IMU located at the heel can provide clinically acceptable estimations of ankle kinematics, implying significant potential for cost and energy savings. Statistical analysis demonstrated the superiority of the developed Long Short-Term Memory (LSTM) network over the other models in intra-subject tasks, achieving impressive accuracy (RMSE: 1.88$\mathrm{^{\circ }}$$\pm$0.02$\mathrm{^{\circ }}$, MAE: 1.41$\mathrm{^{\circ }}$$\pm$0.01$\mathrm{^{\circ }}$, and r2 score: 0.93$\pm$0.01), indicating strong generalization within the same subject. In inter-subject tasks, the convolutional neural network (CNN) and the CNN-LSTM models showed comparable performance but statistically outperformed the other models in terms of estimation accuracy across various inputs. When using a single IMU, the CNN model achieved the lowest error (RMSE: 4.13$\mathrm{^{\circ }}$$\pm$0.55$\mathrm{^{\circ }}$, MAE: 3.33$\mathrm{^{\circ }}$$\pm$0.48$\mathrm{^{\circ }}$, and r2 score: 0.50$\pm$0.21), showcasing its effective generalization to new subjects. Furthermore, deploying the CNN into a microcontroller, with a sinlge IMU at the heel, resulted in promising real-time ankle kinematics estimations (RMSE: 3.34$\mathrm{^{\circ }}$$\pm$0.48$\mathrm{^{\circ }}$, MAE: 2.68$\mathrm{^{\circ }}$$\pm$0.46$\mathrm{^{\circ }}$ and r2 score: 0.63$\pm$0.07). Overall, this research highlights the potential of combining IMUs with ANNs as reliable and practical tools for early prevention and rehabilitation of ankle injuries.

Gait Spatio-Temporal Parameters Vary Significantly Between Indoor, Outdoor and Different Surfaces

Human gait is usually studied in clinical environments, but wearable devices have extended gait analysis beyond traditional assessments. Older adults tend to walk differently indoors and outdoors; however, most gait assessments are conducted on indoor surfaces. It is therefore important to evaluate gait in various outdoor environments. Insights gained from these assessments significantly enhance our understanding of the impact of environmental factors on gait performance and ensure that clinical evaluations are effectively aligned with everyday locomotion. A total of 100 participants with foot pain, 38 young (18-45 years) and 62 older adults (65-80 years), completed a 10-Metre Walk Test (10MWT) in three randomised conditions at their typical, comfortable walking pace, including (1) 10MWT of indoor walking, (2) 10MWT of outdoor walking on grass and (3) 10MWT of outdoor walking on a sidewalk. Wearable inertial sensors recorded gait data and the magnitudes of the following gait measures: gait speed, cadence, stride length, stride duration and asymmetry. A statistical analysis using ANOVA and post hoc comparisons revealed a significantly lower gait speed (p < 0.001), lower stride length (p < 0.001) and lower asymmetry (p < 0.001) indoors compared to outdoors, demonstrating that environmental factors significantly affect spatio-temporal gait parameters. Wearable sensor-based gait analysis performed in controlled clinical settings may underestimate real-life conditions. Some important spatio-temporal parameters, useful in detecting people with gait impairment and at risk of falling, are significantly affected by environment and individual postural ability more than demographic factors.

Beyond boundaries: a location-based toolkit for quantifying group dynamics in diverse contexts

Existing toolkits for analyzing movement dynamics in animal ecology primarily focus on individual or group behavior in habitats without predefined boundaries, while methods for studying human activity often cater to bounded environments, such as team sports played on defined fields. This leaves a gap in tools for modeling and analyzing human group dynamics in large-scale, unbounded, or semi-constrained environments. Examples of such contexts include tourist groups, cycling teams, search and rescue teams, and military units. To address this issue, we survey existing methods and metrics for characterizing individual and collective movement in humans and animals. Using a rich GPS dataset from groups of military personnel engaged in a foot march, we develop a comprehensive, general-purpose toolkit for quantifying group dynamics using location-based metrics during goal-directed movement in open environments. This toolkit includes a repository of Python functions for extracting and analyzing movement data, integrating cognitive factors such as decision-making, situational awareness, and group coordination. By extending location-based analytics to non-traditional domains, this toolkit enhances the understanding of collective movement, group behavior, and emergent properties shaped by cognitive processes. To demonstrate its practical utility, we present a use case utilizing metrics derived from the foot march data to predict group performance during a subsequent strategic and tactical exercise, highlighting the influence of cognitive and decision-making behaviors on team effectiveness.

Impact of smartphone distraction and alcohol intoxication on pedestrian risk-taking

BACKGROUND

An estimated 7,388 pedestrians died in motor-vehicle crashes in the United States in 2021. Two significant risks for pedestrian injuries and deaths are alcohol intoxication and smartphone distraction. The present research used a virtual reality simulator to evaluate the individual and joint impact of pedestrian distraction and intoxication on risk-taking while crossing the street.

METHODS

Thirty-nine participants completed two laboratory visits, during which they crossed the virtual street either after drinking alcohol to produce a BAC of 0.08 or after drinking a placebo, in randomized order. During each visit, they crossed the street both while distracted by texting and without distraction, also in randomized order. Five pedestrian safety outcomes were considered: unsafe crossings, time to contact with oncoming vehicles, start gap before entering a safe gap in traffic, distance to the closest oncoming vehicle as the crossing started, and missed opportunities to cross safely.

RESULTS

Intoxicated participants were more likely to cross unsafely. While distracted, participants missed more safe crossing opportunities, started crossing while closer to oncoming vehicles, and experienced more unsafe crossings. The interactional effect of intoxication and distraction was significant for the number of unsafe crossings and time to contact, with intoxicated pedestrians experiencing more unsafe crossings only when they were not distracted and distraction increasing unsafe crossings only among sober pedestrians.

CONCLUSIONS

Both alcohol intoxication and smartphone distraction impacted pedestrian safety, individually and jointly. Results should inform the development of multifaceted prevention strategies, including road engineering, law enforcement, and efforts to reduce pedestrian risk through strategies like responsible beverage service practices.

PRACTICAL APPLICATIONS

Overall, this study explored the isolated and interactional effects of alcohol intoxication and phone distraction on pedestrians, which should inform development of interventions to reduce risky pedestrian behavior and address pedestrian injury and mortality rates globally.

Are Suggested Hiking Times Accurate? A Validation of Hiking Time Estimations for Preventive Measures in Mountains

Background and Objectives: Accurate hiking time estimate is crucial for outdoor activity planning, especially in mountainous terrains. Traditional mountain signage and online platforms provide generalized hiking time recommendations, often lacking personalization. This study aims to evaluate the variability in hiking time estimates from different methods and assess the potential of a novel algorithm, MOVE, to enhance accuracy and safety. Materials and Methods: A cross-sectional analysis was conducted using data from 25 Italian loop trails selected via the Wikiloc platform, considering user-uploaded GPS data from at least 20 users per trail. Real-world hiking times were compared with estimations from Komoot, Outdooractive, mountain signage, and the MOVE algorithm, which incorporates individualized biological and trail characteristics. Results: Significant discrepancies were observed between actual hiking times and estimates from Komoot (ΔWK: -48.92 ± 57.16 min), Outdooractive (ΔWO: -69.13 ± 58.23 min), and mountain signage (ΔWS: -29.59 ± 59.90 min; all p < 0.001). In contrast, MOVE showed no statistically significant difference (ΔWM: -0.27 ± 65.72 min; p = 0.278), providing the most accurate predictions. Conclusions: Current hiking time estimation methods show substantial variability and inaccuracy, which may pose safety risks. MOVE demonstrated superior accuracy, offering personalized hiking time predictions based on user-specific data and trail characteristics. Integrating such advanced tools into outdoor activity planning could enhance safety and accessibility, particularly for individuals with chronic conditions. Further studies should explore integrating real-time health data to refine these tools.

Comparison of Empirical and Reinforcement Learning (RL)-Based Control Based on Proximal Policy Optimization (PPO) for Walking Assistance: Does AI Always Win?

The use of wearable assistive devices is growing in both industrial and medical fields. Combining human expertise and artificial intelligence (AI), e.g., in human-in-the-loop-optimization, is gaining popularity for adapting assistance to individuals. Amidst prevailing assertions that AI could surpass human capabilities in customizing every facet of support for human needs, our study serves as an initial step towards such claims within the context of human walking assistance. We investigated the efficacy of the Biarticular Thigh Exosuit, a device designed to aid human locomotion by mimicking the action of the hamstrings and rectus femoris muscles using Serial Elastic Actuators. Two control strategies were tested: an empirical controller based on human gait knowledge and empirical data and a control optimized using Reinforcement Learning (RL) on a neuromuscular model. The performance results of these controllers were assessed by comparing muscle activation in two assisted and two unassisted walking modes. Results showed that both controllers reduced hamstring muscle activation and improved the preferred walking speed, with the empirical controller also decreasing gastrocnemius muscle activity. However, the RL-based controller increased muscle activity in the vastus and rectus femoris, indicating that RL-based enhancements may not always improve assistance without solid empirical support.

Estimation bias and serial dependence in speed perception

Studies have found that feature estimates are systematically compressed towards the distribution center, showing a central tendency. Additionally, the estimate of current features is affected by the previously seen feature, showing serial dependence or adaptation effect. However, these all remain unclear in the speed estimation. To address this question, we asked participants to estimate the speed of moving Gabor patches. In Experiment 1, speeds were selected from three uniform distributions with different lower and upper boundaries (i.e., slow, moderate, and fast ranges). In Experiment 2, speeds were arranged in an increasing, uniform, or decreasing distribution. The boundaries of three distributions were the same. The results found that speed estimates were systematically compressed towards the center of the uniform distribution center, showing a central tendency, and its size increased with the range boundaries. However, in the decreasing and increasing distributions, aside from central tendency, the speed estimates were also showed a bias away from the heavy tail of the distributions. Moreover, there was an attractive serial dependence that was not affected by the speed range. In summary, the current study, along with previous studies that reveal a slow-speed bias, comprehensively reveals various estimation biases in speed perception.

Methods for Evaluating Tibial Accelerations and Spatiotemporal Gait Parameters during Unsupervised Outdoor Movement

The purpose of this paper is to introduce a method of measuring spatiotemporal gait patterns, tibial accelerations, and heart rate that are matched with high resolution geographical terrain features using publicly available data. These methods were demonstrated using data from 218 Marines, who completed loaded outdoor ruck hikes between 5-20 km over varying terrain. Each participant was instrumented with two inertial measurement units (IMUs) and a GPS watch. Custom code synchronized accelerometer and positional data without a priori sensor synchronization, calibrated orientation of the IMUs in the tibial reference frame, detected and separated only periods of walking or running, and computed acceleration and spatiotemporal outcomes. GPS positional data were georeferenced with geographic information system (GIS) maps to extract terrain features such as slope, altitude, and surface conditions. This paper reveals the ease at which similar data can be gathered among relatively large groups of people with minimal setup and automated data processing. The methods described here can be adapted to other populations and similar ground-based activities such as skiing or trail running.

A clinical application of gait quality patterns in osteoarthritis

AIM

To investigate whether a smartphone-based gait analysis tool can reliably output gait quality parameters that can be cross-analyzed to establish individual & disease-based changes in gait quality patterns.

METHODS

A cross-sectional study made up of a 48-patients undergoing disability certification at the "Dr. José Castro Villagrana" or the "Dr. David Fragoso Lizalde" Health Centers in Mexico City, Mexico. Their sensorimotor performance was evaluated through an in-house smartphone/IMU based digital tool. Gait was analyzed by means of frequency analysis of the acceleration of the body mass measured at the sternum. A composite gait quality score was determined through principal component analysis based primarily on the explainability and uniformity of gait. Quality independence against demographic variables (age & weight) was tested through ANCOVA. The association between gait quality and gait parameters was analyzed by using multiple linear regression.

RESULTS

A multiple regression model developed with a limited set of gait quality parameters successfully predicted gait smoothness with a 97.05 % accuracy with a mean square error of 0.085 between predicted and actual quality scores. The model demonstrates different predictive capacities across disease groups, with Osteoarthrosis + Osteoporosis having the highest R2 at 0.98 (p < 0.001) and Coxarthrosis having the lowest explained R2 at 0.79 (p < 0.001).

CONCLUSIONS

The assessment of gait quality, in family medicine, with low-cost digital tools is an area of opportunity yet to be explored. This tool can potentially disrupt the current disability workflow between primary and specialty care to have an objective method of assessing gait within a clinical consult. Individual patient-level benchmarking can give us insights into the patient's disease status, develop practical intervention strategies, and control the cost and quality of medical care by predicting an individualized course of disability or rehabilitation. Further studies are needed to validate digital gait assessments as clinical decision support tools for day-to-day clinical operations. MESH: Gait Analysis, Smartphone, Primary Health Care, Osteoarthrosis.

Reframing Whole-Body Angular Momentum: Exploring the Impact of Low-Pass Filtered Dynamic Local Reference Frames During Straight-Line and Turning Gaits

Accurately estimating whole-body angular momentum (WBAM) during daily activities may benefit from choosing a locally-defined reference frame aligned with anatomical axes, particularly during activities involving body turns. Local reference frames, potentially defined by pelvis heading angles, horizontal center of mass velocity (vCoM), or average angular velocity ( Aω ), can be utilized. To minimize the impact of inherent mediolateral oscillations of these frames, such as those caused by pelvis or vCoM rotation in the transverse plane, a low-pass filter is recommended. This study investigates how differences among global, local reference frames pre- and post-filtering affect WBAM component distribution across anatomical axes during straight-line walking and various turning tasks, which is lacking in the literature. Results highlighted significant effects of reference frame choice on WBAM distribution in the anteroposterior (AP) and mediolateral (ML) axes in all tasks. Specifically, expressing WBAM in the vCoM-oriented local reference frame yielded significantly lower (or higher) WBAM in the AP (or ML) axes compared to pelvis-oriented and Aω -oriented frames. However, these significant differences disappeared after employing a low-pass filter to local reference frames. Therefore, employing low-pass filtered local reference frames is crucial to enhance their applicability in both straight-line and turning tasks, ensuring more precise WBAM estimates. In applications that require expressing anatomical axes-dependent biomechanical parameters in a local reference frame, pelvis- and vCoM-oriented frames are more practical compared to the A ω -oriented frame, as they can be determined by a reduced optical marker set or inertial sensors in future applications when the whole-body kinematics is not available.

Referent Control of Side-to-Side Body-Weight Transfer During Standing and Stepping in Adults

Research suggests that locomotion may be primarily caused by shifting stable body balance from one location in the environment to another with subsequent rhythmical muscle activation by the central pattern generator (CPG), constituting a multi-level control system. All levels interact with environmental forces affected by proprioceptive and vestibular reflexes as well as vision. A similar multi-level control schema is likely used to shift body balance laterally when the body weight is rhythmically transferred from side-to-side. In order to do so, the system shifts a specific body posture in space. This body posture is referred to as the threshold or referent body posture, R, at which all muscles involved can be at rest but are activated depending on the deflection of the actual body posture, Q, from R. This concept has previously been investigated for forward and backward locomotion. The purpose of the present study was to verify if it was also applicable to locomotor tasks in other directions such as sidestepping. We predicted that during sidestepping, the actual and referent posture can transiently match each other bringing the activity of multiple muscles to a minimum. The existence of such minima was demonstrated in healthy adults performing three locomotor tasks involving shifts of the body weight from side-to-side thus further supporting the validity of the multi-level control scheme of locomotion.

Potential exposure of adults and children to particles from resuspended nano-enabled consumer sprays

The increasing application of nanotechnology has resulted in a growing number of nano-enabled consumer products, and they could be important contributors to indoor particulate matter, with potential adverse health effects. This study investigated the exposure of adults and children to the released and resuspended manufactured particles from seven nano-enabled consumer sprays. Sedimentation and resuspension of released particles were investigated in a newly constructed 2.8 × 1.6 × 2.4 m3 chamber. The resuspension of deposited particles was investigated as a function of product type, flooring material (e.g., carpet and vinyl), resuspension force (e.g., walking by an adult and motion of a robotic sampler that simulated a child), and measurement height. The concentration of released and resuspended particles in the air was determined using Button Aerosol Samplers (SKC Inc.) with 25-mm 2 μm-pore PTFE filters. Samplers were positioned in the experimenter's breathing zone (e.g., 1.5 m for adults and 0.3 m for a child-simulating robot) and at fixed stations of 0.3 m and 1.1 m heights. Resuspended particle mass concentrations ranged from 28 to 905 μg/m3, and the resuspension rates of deposited spray particles for the same variable combinations varied from 10-4 to 10-1 h-1, depending on product type, flooring material, sampling height, and resuspension force. Particle resuspension rates from carpet were up to 320 % higher than resuspension rates from vinyl flooring, resuspension rates measured at 0.3 m were up to 195 % higher than the rates measured with a 1.1 m stationary sampler, and resuspension rates due to a walking adult were up to 243 % higher than resuspension rates caused to a moving robot that simulated a child. Overall, these data on the resuspension of particles from nano-enabled consumer sprays could help us understand the resulting exposures and support future studies on human exposure reduction.

The Effectiveness and Perceptions of Three Moderate Intensity Walking Cadence Aids and their Effects on Affective States: A Mixed Methods Study

Substantial health benefits can be derived from walking at a moderate intensity cadence. To help regulate this cadence, three distinct aids exist 1) self-perception; 2) cadence prescription; 3) auditory cues. This study aimed to compare the effectiveness and explore perceptions of these aids to promote moderate intensity walking and effects on affective states, thereby addressing an important research gap. Individualised moderate relative intensity waking cadence was determined for participants (n = 23, Mage = 26.35, SD = 10.11). A convergent mixed-methods design was employed. A within-persons repeated measures design was used to explore the effectiveness of three aids (general guidelines, cadence prescription, and music) on promoting moderate intensity physical activity and positive affective states. Perceptions of these aids were elicited through qualitative interviews and thematic content analysis. Main effects for condition on relative physical activity intensity (η2 = .72) and positive affect (η2 = .25) were observed. Music evoked significantly higher relative physical activity intensity than other conditions (p values < .01), and higher positive affect compared to the general guidelines condition (p = .038). A significantly greater proportion of participants achieved moderate relative intensity physical activity during the music compared to general guidelines condition (p = .03). Congruently, qualitative findings suggested that participants predominantly perceived music as most effective for promoting a moderate intensity cadence and positive affect. However, individual variability existed in ability to utilise this aid. Implications of the findings for practitioners seeking to promote a moderate intensity cadence and positive affect during walking are discussed.

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.

Automated external defibrillator location and socioeconomic deprivation in Great Britain

OBJECTIVE

The early use of automated external defibrillators (AEDs) improves outcomes in out-of-hospital cardiac arrest (OHCA). We investigated AED access across Great Britain (GB) according to socioeconomic deprivation.

METHODS

Cross-sectional observational study using AED location data from The Circuit: the national defibrillator network led by the British Heart Foundation in partnership with the Association of Ambulance Chief Executives, Resuscitation Council UK and St John Ambulance. We calculated street network distances between all 1 677 466 postcodes in GB and the nearest AED and used a multilevel linear mixed regression model to investigate associations between the distances from each postcode to the nearest AED and Index of Multiple Deprivation, stratified by country and according to 24 hours 7 days a week (24/7) access.

RESULTS

78 425 AED locations were included. Across GB, the median distance from the centre of a postcode to an AED was 726 m (England: 739 m, Scotland: 743 m, Wales: 512 m). For 24/7 access AEDs, the median distances were further (991 m, 994 m, 570 m). In Wales, the average distance to the nearest AED and 24/7 AED was shorter for the most deprived communities. In England, the average distance to the nearest AED was also shorter in the most deprived areas. There was no association between deprivation and average distance to the nearest AED in Scotland. However, the distance to the nearest 24/7 AED was greater with increased deprivation in England and Scotland. On average, a 24/7 AED was in England and Scotland, respectively, 99.2 m and 317.1 m further away in the most deprived than least deprived communities.

CONCLUSION

In England and Scotland, there are differences in distances to the nearest 24/7 accessible AED between the most and least deprived communities. Equitable access to 'out-of-hours' accessible AEDs may improve outcomes for people with OHCA.

A park-based group mobility program for older adults with difficulty walking outdoors: a quantitative process evaluation of the Getting Older Adults Outdoors (GO-OUT) randomized controlled trial

BACKGROUND

Process evaluations of randomized controlled trials (RCTs) of community exercise programs are important to help explain the results of a trial and provide evidence of the feasibility for community implementation. The objectives of this process evaluation for a multi-centre RCT of outdoor walking interventions for older adults with difficulty walking outdoors, were to determine: 1) implementation fidelity (the extent to which elements of the intervention were delivered as specified in the original protocol) and 2) participant engagement (the receipt of intervention components by the participants) in the Getting Older Adults Outdoors (GO-OUT) trial.

METHODS

GO-OUT participants attended an active 1-day workshop designed to foster safe, outdoor walking skills. After the workshop, 190 people at 4 sites were randomized to an outdoor walk group (OWG) (n = 98) which met 2x/week for 10 weeks, or the weekly reminders (WR) group (n = 92) which received a phone reminder 1x/week for 10 weeks. The OWG had 5 components - warm-up, continuous distance walk, task-oriented walking activities, 2nd continuous distance walk, and cool-down. Data on implementation fidelity and participant engagement were gathered during the study through site communications, use of standardized forms, reflective notes of the OWG leaders, and accelerometry and GPS assessment of participants during 2 weeks of the OWG.

RESULTS

All sites implemented the workshop according to the protocol. Participants were engaged in all 8 activity stations of the workshop. WR were provided to 96% of the participants in the WR intervention group. The 5 components of the OWG sessions were implemented in over 95% of the sessions, as outlined in the protocol. Average attendance in the OWG was not high - 15% of participants did not attend any sessions and 64% of participants in the OWG attended > 50% of the sessions. Evaluations with accelerometry and GPS during week 3 and 9 OWG sessions suggest that participants who attended were engaged and active during the OWG.

CONCLUSIONS

This process evaluation helps explain the main study findings and demonstrates the flexibility required in the protocol for safe and feasible community implementation. Future research could explore the use of additional behaviour change strategies to optimize attendance for community implementation.

TRIAL REGISTRATION

ClinicalTrials.gov NCT03292510 Date of registration: September 25, 2017.

Shared Autonomy Locomotion Synthesis With a Virtual Powered Prosthetic Ankle

Virtual environments provide a safe and accessible way to test innovative technologies for controlling wearable robotic devices. However, to simulate devices that support walking, such as powered prosthetic legs, it is not enough to model the hardware without its user. Predictive locomotion synthesizers can generate the movements of a virtual user, with whom the simulated device can be trained or evaluated. We implemented a Deep Reinforcement Learning based motion controller in the MuJoCo physics engine, where autonomy over the humanoid model was shared between the simulated user and the control policy of an active prosthesis. Despite not optimising the controller to match experimental dynamics, realistic torque profiles and ground reaction force curves were produced by the agent. A data-driven and continuous representation of user intent was used to simulate a Human Machine Interface that controlled a transtibial prosthesis in a non-steady state walking setting. The continuous intent representation was shown to mitigate the need for compensatory gait patterns from their virtual users and halve the rate of tripping. Co-adaptation was identified as a potential challenge for training human-in-the-loop prosthesis control policies. The proposed framework outlines a way to explore the complex design space of robot-assisted gait, promoting the transfer of the next generation of intent driven controllers from the lab to real-life scenarios.

Using Mobile Ecological Momentary Assessment to Understand Consumption and Context Around Online Food Delivery Use: Pilot Feasibility and Acceptability Study

BACKGROUND

Mobile ecological momentary assessment (EMA) is a powerful tool for collecting real-time and contextual data from individuals. As our reliance on online technologies to increase convenience accelerates, the way we access food is changing. Online food delivery (OFD) services may further encourage unhealthy food consumption habits, given the high availability of energy-dense, nutrient-poor foods. We used EMA to understand the real-time effects of OFD on individuals' food choices and consumption behaviors.

OBJECTIVE

The primary aims of this pilot study were to assess the feasibility and acceptability of using EMA in young users of OFD and compare 2 different EMA sampling methods. The secondary aims were to gather data on OFD events and their context and examine any correlations between demographics, lifestyle chronic disease risk factors, and OFD use.

METHODS

This study used EMA methods via a mobile app (mEMASense, ilumivu Inc). Existing users of OFD services aged 16 to 35 years in Australia who had access to a smartphone were recruited. Participants were randomly assigned to 1 of 2 groups: signal-contingent or event-contingent. The signal-contingent group was monitored over 3 days between 7 AM and 10 PM. They received 5 prompts each day to complete EMA surveys via the smartphone app. In contrast, the event-contingent group was monitored over 7 days and was asked to self-report any instance of OFD.

RESULTS

A total of 102 participants were analyzed, with 53 participants in the signal-contingent group and 49 participants in the event-contingent group. Compliance rates, indicating the feasibility of signal-contingent and event-contingent protocols, were similar at 72.5% (574/792) and 73.2% (251/343), respectively. Feedback from the participants suggested that the EMA app was not easy to use, which affected their acceptability of the study. Participants in the event-contingent group were 3.53 (95% CI 1.52-8.17) times more likely to have had an OFD event captured during the study. Pizza (23/124, 18.5%) and fried chicken (18/124, 14.5%) comprised a bulk of the 124 OFD orders captured. Most orders were placed at home (98/124, 79%) for 1 person (68/124, 54.8%). Age (incidence rate ratio 0.95, 95% CI 0.91-0.99; P=.03) and dependents (incidence rate ratio 2.01, 95% CI 1.16-3.49; P=.01) were significantly associated with the number of OFD events in a week after adjusting for gender, socioeconomic status, diet quality score, and perceived stress levels.

CONCLUSIONS

This pilot study showed that EMA using an event-contingent sampling approach may be a better method to capture OFD events and context than signal-contingent sampling. The compliance rates showed that both sampling methods were feasible and acceptable. Although the findings from this study have gathered some insight on the consumption and context of OFD in young people, further studies are required to develop targeted interventions.

Functional popliteal angle tests improve identification of short hamstring muscle-tendon length in patients with a central neurological lesion

This study introduces a functional exercise protocol to improve the identification for short hamstring muscle-tendon length (HMTL), a common contributor to crouch gait in patients with central neurological lesions (CNL). The functional exercise protocol incorporates a knee extension movement with hip in a flexed position, while standing on one leg (functional popliteal angle test) and walking with large steps to the current standard protocol (walking at comfortable speed and as fast as possible). The main aim was to establish whether the new protocol allows better determination of maximum HMTLs and diagnostics of short HMTL in patients with a CNL. Lower limb 3D marker position data from 39 patient limbs and 10 healthy limbs performing the exercises were processed in OpenSim to extract HMTLs. The new protocol provoked significantly larger HMTLs compared to the current standard protocol. The total number of limbs classified as having too short HMTLs reduced from 16 to 4 out of a total of 30 limbs walking in crouch. The new protocol improves determination of maximum HMTL, thereby improving short HMTL diagnostics and identification of patients in need of lengthening treatment. Inter-individual variability observed among patients, indicating the need to include all exercises for comprehensive diagnosis.

High-density EMG, IMU, kinetic, and kinematic open-source data for comprehensive locomotion activities

Novel sensor technology enables new insights in the neuromechanics of human locomotion that were previously not possible. Here, we provide a dataset of high-density surface electromyography (HDsEMG) and high-resolution inertial measurement unit (IMU) signals, along with motion capture and force data for the lower limb of 10 healthy adults during multiple locomotion modes. The participants performed level-ground and slope walking, as well as stairs ascent/descent, side stepping gait, and stand-to-walk and sit-to-stand-to-walk, at multiple walking speeds. These data can be used for the development and validation of locomotion mode recognition and control algorithms for prosthetics, exoskeletons, and bipedal robots, and for motor control investigations.

Computer simulation of scavenging by hominins and giant hyenas in the late Early Pleistocene

Consumption of animal-sourced food is an important factor in broadening the diet of early hominins, promoting brain and body growth, and increasing behavioural complexity. However, whether early hominins obtained animal food by scavenging or hunting large mammals remains debated. Sabre-toothed felids have been proposed to facilitate the expansion of early Homo out of Africa into Europe 1.4-0.8 Ma by creating a niche for scavengers in Eurasia as the carcasses abandoned by these felids still contained abundant edible resources. In contrast, it has been argued that the niche for a large scavenger was already occupied in Eurasia by the giant hyena, preventing hominins from utilising this resource. This study shows that sabre-toothed felids generated carcasses rich in edible resources and that hominins were capable of competing with giant hyenas for this resource. The simulation experiments showed that maintaining an optimum group size is essential for the success of the hominin scavenging strategy. Early hominins could outcompete giant hyenas only if they could successfully dispute carcasses with them. Thus, in the presence of a strong competitor, passive scavenging is essentially the same as confrontational scavenging.

Pedestrian Localization with Stride-Wise Error Estimation and Compensation by Fusion of UWB and IMU Data

Indoor positioning enables mobile machines to perform tasks (semi-)automatically, such as following an operator. However, the usefulness and safety of these applications depends on the reliability of the estimated operator localization. Thus, quantifying the accuracy of positioning at runtime is critical for the application in real-world industrial contexts. In this paper, we present a method that produces an estimate of the current positioning error for each user stride. To accomplish this, we construct a virtual stride vector from Ultra-Wideband (UWB) position measurements. The virtual vectors are then compared to stride vectors from a foot-mounted Inertial Measurement Unit (IMU). Using these independent measurements, we estimate the current reliability of the UWB measurements. Positioning errors are mitigated through loosely coupled filtering of both vector types. We evaluate our method in three environments, showing that it improves positioning accuracy, especially in challenging conditions with obstructed line of sight and sparse UWB infrastructure. Additionally, we demonstrate the mitigation of simulated spoofing attacks on UWB positioning. Our findings indicate that positioning quality can be judged at runtime by comparing user strides reconstructed from UWB and IMU measurements. Our method is independent of situation- or environment-specific parameter tuning, and as such represents a promising approach for detecting both known and unknown positioning error states.

Different types of plantar vibration affect gait characteristics differently while walking on different inclines

Background

Plantar vibration has been widely used to strengthen the sensation of the somatosensory system, further enhancing balance during walking on a level surface in patients with stroke. However, previous studies with plantar vibration only involved the level surface, which neglected the importance of inclined/declined walking in daily life. Thus, combining the plantar vibration and inclined/declined walking might answer a critical research question: whether different types of plantar vibration had different effects on gait characteristics during walking on different inclines.

Methods

Eighteen healthy young adults were recruited. Fifteen walking conditions were assigned randomly to these healthy adults (no, sub-, and supra-threshold plantar vibration × five different inclines: +15%, +8%, 0%, -8%, -15% grade). A motion capture system with eight cameras captured 12 retro-reflective markers and measured the stride time, stride length, step width, and respective variabilities.

Results

A significant interaction between vibration and inclination was observed in the stride time (p < 0.0001) and step width (p = 0.015). Post hoc comparisons found that supra-threshold vibration significantly decreased the stride time (-8%: p < 0.001; -15%: p < 0.001) while the sub-threshold vibration significantly increased the step width (-8%: p = 0.036) in comparison with no plantar vibration.

Conclusions

When walking downhill, any perceivable (supra-threshold) vibration on the plantar area decreased the stride time. Also, the increase in step width was observed by non-perceivable (sub-threshold) plantar vibration while walking uphill. These observations were crucial as follows: (1) applying sub-threshold plantar vibrations during uphill walking could increase the base of support, and (2) for those who may need challenges in locomotor training, applying supra-threshold vibration during downhill walking could reach this specific training goal.

Effect of Treadmill Training with Visual Biofeedback on Selected Gait Parameters in Subacute Hemiparetic Stroke Patients

BACKGROUND

Functional limitations after a stroke are unique to each person and often include impaired independent mobility. A reduction in existing gait deficits after a stroke is often one of the main goals of rehabilitation. Gait re-education after stroke is a complex process, which consists of the effects of many therapeutic interventions.

OBJECTIVE

The study aimed to analyze the effects of using a treadmill with visual feedback in gait re-education in the sub-acute stroke period and assess the impact of biofeedback treadmill training on selected gait parameters, improving static balance and reducing the need for orthopedic aids.

METHODS

The study included 92 patients (F: 45, M: 47) aged 63 ± 12 years, with post-ischemic sub-acute (within six months onset) stroke hemiparesis, treated at a neurological rehabilitation ward. All patients participated in a specific rehabilitation program, and in addition, patients in the study group (n = 62) have a further 10 min of treadmill training with visual feedback. Patients in the control group (n = 30) participated in additional conventional gait training under the direct supervision of a physiotherapist. The evaluation of static balance was assessed with the Romberg Test. A Biodex Gait Trainer 3 treadmill with biofeedback function was used to evaluate selected gait parameters (walking speed, step length, % limb loading, and traveled distance). The use of an orthopedic aid (walker or a crutch) was noted.

RESULTS

After four weeks of rehabilitation, step length, walking speed, traveled distance, and static balance were significantly improved for the study and control group (p < 0.05). Treadmill gait training yielded significantly better results than a conventional rehabilitation program. Only the study group observed a corrected walking base (p < 0.001). All participants showed a reduction in the use of walking aids (p = 0.006). There was no asymmetry in the % of limb loading for either group prior to or following rehabilitation.

CONCLUSIONS

The treadmill with visual biofeedback as conventional gait training has resulted in a significant improvement in parameters such as step length, walking speed, static balance, and a reduction in the use of locomotion aids. However, the achieved improvement in gait parameters is still not in line with the physiological norm.

Slow and Steady But Not Related to HIV Stigma: Physical Activity in South Africans Living with HIV and Chronic Pain

HIV stigma may influence physical activity in people living with HIV (PLWH) and chronic pain. We prospectively examined the relationship between stigma, activity and chronic pain in a convenience sample of PLWH initiating antiretroviral therapy in an inner-city clinic in Johannesburg, South Africa. Participants wore accelerometers to measure daily duration and intensity of activity for 2 weeks. Stigma was assessed with the Revised HIV Stigma Scale. Participants [n = 81, 89% female, age mean (SD) 42 (8)] were active for a median of 7 h daily (IQR 5.2, 9.2), but at very low intensity, equivalent to a slow walk [median (IQR): 0.39 m s^-1 (0.33, 0.50)]. Duration and intensity of activity was not associated with stigma, even after controlling for age, self-assessed wealth, pain intensity and willingness to engage in physical activity (p-values > 0.05). As stigma did not associate with greater activity, drivers of sustained activity in South African PLWH remain unclear.

Unsupervised Early Detection of Physical Activity Behaviour Changes from Wearable Accelerometer Data

Wearable accelerometers record physical activity with high resolution, potentially capturing the rich details of behaviour changes and habits. Detecting these changes as they emerge is valuable information for any strategy that promotes physical activity and teaches healthy behaviours or habits. Indeed, this offers the opportunity to provide timely feedback and to tailor programmes to each participant's needs, thus helping to promote the adherence to and the effectiveness of the intervention. This article presents and illustrates U-BEHAVED, an unsupervised algorithm that periodically scans step data streamed from activity trackers to detect physical activity behaviour changes to assess whether they may become habitual patterns. Using rolling time windows, current behaviours are compared with recent previous ones, identifying any significant change. If sustained over time, these new behaviours are classified as potentially new habits. We validated this detection algorithm using a physical activity tracker step dataset (N = 12,798) from 79 users. The algorithm detected 80% of behaviour changes of at least 400 steps within the same hour in users with low variability in physical activity, and of 1600 steps in those with high variability. Based on a threshold cadence of approximately 100 steps per minute for standard walking pace, this number of steps would suggest approximately 4 and 16 min of physical activity at moderate-to-vigorous intensity, respectively. The detection rate for new habits was 80% with a minimum threshold of 500 or 1600 steps within the same hour in users with low or high variability, respectively.

Perspectives on exercise intensity, volume, step characteristics and health outcomes in walking for transport

Background

Quantification of movement intensity and energy utilization, together with frequency of trips, duration, distance, step counts and cadence, is essential for interpreting the character of habitual walking for transport, and its potential support of health. The purpose of the study is to illuminate this with valid methods and novel perspectives, and to thereby provide a new basis for characterizing and interpreting walking in relation to health outcomes.

Methods

Habitual middle-aged commuting pedestrians (males = 10, females = 10) were investigated in the laboratory at rest and with maximal treadmill and cycle ergometer tests. Thereafter, levels of oxygen uptake, energy expenditure, ventilation, heart rate, blood lactate, rated perceived exertion, cadence, number of steps, duration, distance, and speed were recorded during the normal walking commute of each participant in Greater Stockholm, Sweden. The number of commutes per week over the year was self-reported.

Results

Walking in the field demanded about 30% more energy per km compared to level treadmill walking. For both sexes, the walking intensity in field was about 46% of maximal oxygen uptake, and energy expenditure amounted to 0.96 kcal · kg- 1 · km- 1. The MET values (males: 6.2; females: 6.5) mirrored similar levels of walking speed (males: 5.7; females: 5.9 km · h- 1) and levels of oxygen uptake (males: 18.6; females: 19.5 mL · kg- 1 · min- 1). The average number of MET-hours per week in a typical month was 22 for males and 20 for females. This resulted in a total weekly energy expenditure of ~1,570 and 1,040 kcal for males and females, respectively. Over the year, the number of walking commutes and their accumulated distance was ~385 trips and 800 km for both sexes.

Conclusion

Walking in naturalistic field settings demands its own studies. When males and females walk to work, their relative aerobic intensities and absolute energy demands for a given distance are similar. It is equivalent to the lower part of the moderate relative intensity domain. The combination of oxygen uptake, trip duration and frequency leads to high and sustained levels of MET-hours as well as energy expenditure per week over the year, with a clear health enhancing potential. Based on this study we recommend 6000 transport steps per day, or equivalent, during five weekdays, over the year, in order to reach optimal health gains.

Using open data and open-source software to develop spatial indicators of urban design and transport features for achieving healthy and sustainable cities

Benchmarking and monitoring of urban design and transport features is crucial to achieving local and international health and sustainability goals. However, most urban indicator frameworks use coarse spatial scales that either only allow between-city comparisons, or require expensive, technical, local spatial analyses for within-city comparisons. This study developed a reusable, open-source urban indicator computational framework using open data to enable consistent local and global comparative analyses. We show this framework by calculating spatial indicators-for 25 diverse cities in 19 countries-of urban design and transport features that support health and sustainability. We link these indicators to cities' policy contexts, and identify populations living above and below critical thresholds for physical activity through walking. Efforts to broaden participation in crowdsourcing data and to calculate globally consistent indicators are essential for planning evidence-informed urban interventions, monitoring policy effects, and learning lessons from peer cities to achieve health, equity, and sustainability goals.

The effects of custom-made foot orthoses on foot pain, foot function, gait function, and free-living walking activities in people with psoriatic arthritis (PsA): a pre-experimental trial

Introduction

Foot involvement is a significant concern in psoriatic arthritis (PsA) as it can lead to severe levels of foot pain and disability and reduced mobility and quality of life. Previous studies have shown moderate efficacy for custom-made foot orthoses (CFO) in reducing foot pain and disability in people with rheumatoid arthritis. However, evidence on the efficacy of CFO in people with PsA is lacking.

Objectives

To explore the effects of CFO on foot function, foot and lower limb pain, gait function, and free-living walking activities (FWA) in people with PsA.

Methods

A pre-experimental study including twenty participants with PsA (mean age: 54.10 ± 9.06 years and disease duration: 11.53 ± 10.22 years) was carried out. All the participants received and wore CFO for 7 weeks. Foot and lower limb pain and foot function were measured before and after the intervention using the numerical rating scale (NRS) and the foot function index (FFI). Gait function was assessed by recording spatiotemporal parameters (STPs) during a 10-m walk test using an instrumented gait analysis system (Mobility Lab). Free-living walking activities (step count, free-living cadence, time spent in different ambulatory physical activities (APA)) were recorded over 7 days using an accelerometer-instrumented sock.

Results

The FFI reported scores demonstrated severe baseline levels of foot pain (54.46 ± 14.58 %) and disability (46.65 ± 16.14%). Statistically and clinically significant improvements in foot pain and foot function and large effect sizes (Cohen’s effect size > 1, p < 0.005) were observed after the intervention period. A strong correlation (r = −0.64, p < 0.01) between the CFO wearing time and foot function was demonstrated. However, no significant changes were found for gait STP or free-living walking activities after 7 weeks of CFO use.

Conclusion

Results support the clinical and biomechanical plausibility of using CFO in people with PsA to reduce pain and improve foot function. Large-scale and controlled studies are needed to confirm these findings. Moreover, a multidisciplinary approach including the prescription of exercise therapy and physiotherapy combined with CFO could be required to improve STP and promote APA in people with PsA.

Trial registration

ClinicalTrials.gov, NCT05075343. Retrospectively registered on September 29, 2021

Supplementary Information

The online version contains supplementary material available at 10.1186/s13075-022-02808-8.

Exoskeletons and Exosuits Could Benefit from Mode-Switching Body Interfaces That Loosen/Tighten to Improve Thermal Comfort

Exoskeletons and exosuits (exos) are wearable devices that physically assist movement. User comfort is critically important for societal adoption of exos. Thermal comfort (a person's satisfaction with their thermal environment) represents a key design challenge. Exos must physically attach/interface to the body to apply forces, and these interfaces inevitably trap some heat. It is envisioned that thermal comfort could be improved by designing mode-switching exo interfaces that temporarily loosen around a body segment when assistive forces are not being applied. To inform exo design, a case series study (N = 4) based on single-subject design principles was performed. Our objective was to assess individual responses to skin temperature and thermal comfort during physical activity with a Loose leg-sleeve interface compared with a Form-Fitting one, and immediately after a Form-Fitting sleeve switched to Loose. Skin under the Loose sleeve was 2-3 °C (4-6 °F) cooler after 25 min of physical activity, and two of four participants reported the Loose sleeve improved their thermal comfort. After completion of the physical activity, the Form-Fitting sleeve was loosened, causing a 2-4 °C (3-8 °F) drop in skin temperature underneath for all participants, and two participants to report slightly improved thermal comfort. These findings confirmed that an exo that can quickly loosen its interface when assistance is not required-and re-tighten when it is- has the potential to enhance thermal comfort for some individuals and environments. More broadly, this study demonstrates that mode-switching mechanisms in exos can do more than adjust physical assistance: they can also exploit thermodynamics and facilitate thermoregulation in a way that enhances comfort for exo users.

Real-world gait speed estimation, frailty and handgrip strength: a cohort-based study

Gait speed is a reliable outcome measure across multiple diagnoses, recognized as the 6th vital sign. The focus of the present study was on assessment of gait speed in long-term real-life settings with the aim to: (1) demonstrate feasibility in large cohort studies, using data recorded with a wrist-worn accelerometer device; (2) investigate whether the walking speed assessed in the real-world is consistent with expected trends, and associated with clinical scores such as frailty/handgrip strength. This cross-sectional study included n = 2809 participants (1508 women, 1301 men, [45-75] years old), monitored with a wrist-worn device for 13 consecutive days. Validated algorithms were used to detect the gait bouts and estimate speed. A set of metrics were derived from the statistical distribution of speed of gait bouts categorized by duration (short, medium, long). The estimated usual gait speed (1-1.6 m/s) appears consistent with normative values and expected trends with age, gender, BMI and physical activity levels. Speed metrics significantly improved detection of frailty: AUC increase from 0.763 (no speed metrics) to 0.798, 0.800 and 0.793 for the 95th percentile of individual's gait speed for bout durations < 30, 30-120 and > 120 s, respectively (all p < 0.001). Similarly, speed metrics also improved the prediction of handgrip strength: AUC increase from 0.669 (no speed metrics) to 0.696, 0.696 and 0.691 for the 95th percentile of individual's gait speed for bout durations < 30, 30-120 and > 120 s, respectively (all p < 0.001). Forward stepwise regression showed that the 95th percentile speed of gait bouts with medium duration (30-120 s) to be the best predictor for both conditions. The study provides evidence that real-world gait speed can be estimated using a wrist-worn wearable system, and can be used as reliable indicator of age-related functional decline.

Music Tempo: A Tool for Regulating Walking Cadence and Physical Activity Intensity in Overweight Adults?

This study investigated if music tempo can prompt a desired walking cadence, and if music can provide a stimulus to regulate physical activity intensity in a longitudinal physical activity intervention with free-living adults. Overweight adults (n = 37; 94.26 ± 17.11 kg; 49.63 ± 12.37 years) were randomly assigned to an intervention (IG, n = 17) or usual care group (UC, n = 20) as part of a novel nine-month walking intervention. IG participants walked to self-selected music with a predetermined tempo and received a behavioural change support programme. At baseline, four-, six- and nine-months participants were asked to walk around an elliptical track at their habitual pace (0-2 min) and then in time to a predetermined tempo (2-8 min) designed to elicit moderate intensity. Cadence response (steps/min) was assessed and intensity (heart rate (bpm) recorded using wireless telemetry. A repeated measures general linear model (GLM) examined differences between groups over time (p < 0.05). All data is presented as means ± SD. At each assessment point both groups displayed an immediate cadence adjustment in response to music tempo (p < 0.01) i.e., habitual cadence vs. 3 METs target cadence (p < 0.05) and 3 METs target cadence vs. 5 METs target cadence (p < 0.05). Additionally, IG participants displayed an increased habitual cadence (0-2 min) at each assessment point (p < 0.05; 110 ± 9, 121.80 ± 7.5, 121.46 ± 10, 121.93 ± 7 steps/min respectively). UC participant's habitual cadence was unchanged from 0-9 months (p > 0.05; 120 ± 10, 116 ± 13, 119 ± 12 and 119 ± 9 steps/min respectively). Music tempo may be a useful regulatory tool to prompt the free-living individual to reach an appropriate stride rate to achieve a walking pace that is at least moderate intensity. It also appears that results may be trainable as throughout the study an increased habitual walking cadence was observed, in the absence of music.

Real-time conversion of inertial measurement unit data to ankle joint angles using deep neural networks

Joint angle quantification from inertial measurement units (IMUs) is commonly performed using kinematic modelling, which depends on anatomical sensor placement and/or functional joint calibration; however, accurate three-dimensional joint motion measurement remains challenging to achieve. The aims of this study were firstly to employ deep neural networks to convert IMU data to ankle joint angles that are indistinguishable from those derived from motion capture-based inverse kinematics (IK) - the reference standard; and secondly, to validate the robustness of this approach across contrasting walking speeds in healthy individuals. Kinematics data were simultaneously calculated using IMUs and IK from 9 subjects during walking on a treadmill at 0.5 m/s, 1.0 m/s and 1.5 m/s. A generative adversarial network was trained using gait data at two of the walking speeds to predict ankle kinematics from IMU data alone for the third walking speed. There were significant differences between IK and IMU joint angle predictions for ankle eversion and internal rotation during walking at 0.5 m/s and 1.0 m/s (p < 0.001); however, no significant differences in joint angles were observed between the generative adversarial network prediction and IK at any speed or plane of joint motion (p < 0.05). The RMS difference in ankle joint kinematics between the generative adversarial network and IK for walking at 1.0 m/s was 3.8°, 2.1° and 3.5° for dorsiflexion, inversion and axial rotation, respectively. The modeling approach presented for real-time IMU to ankle joint angle conversion, which can be readily expanded to other joints, may provide enhanced IMU capability in applications such as telemedicine, remote monitoring and rehabilitation.

Measuring Outdoor Walking Capacities Using Global Positioning System in People with Multiple Sclerosis: Clinical and Methodological Insights from an Exploratory Study

We aimed at showing how Global Positioning System (GPS) along with a previously validated speed processing methodology could be used to measure outdoor walking capacities in people with multiple sclerosis (MS). We also deal with methodological issues that may occur when conducting such measurements, and explore to what extent GPS-measured outdoor walking capacities (maximal walking distance [MWD_GPS] and usual walking speed) could be related to traditional functional outcomes (6-min total walking distance) in people with MS. Eighteen people with MS, with an Expanded Disability Status Scale score ≤6, completed a 6-min walking test and an outdoor walking session (60 min maximum) at usual pace during which participants were wearing a DG100 GPS receiver and could perform several walking bouts. Among the 12 participants with valid data (i.e., who correctly completed the outdoor session with no spurious GPS signals that could prevent the detection of the occurrence of a walking/stopping bout), the median (90% confidence interval, CI) outdoor walking speed was 2.52 km/h (2.17; 2.93). Ten participants (83% (56; 97)) had ≥1 stop during the session. Among these participants, the median of MWD_GPS was 410 m (226; 1350), and 40% (15; 70) did not reach their MWD_GPS during the first walking bout. Spearman correlations of MWD_GPS and walking speed with 6-min total walking distance were, respectively, 0.19 (-0.41; 0.95) and 0.66 (0.30; 1.00). Further work is required to provide guidance about GPS assessment in people with MS.