Search Results (74)

Background/Objectives: Hajj is a major annual mass gathering. It requires prolonged walking under conditions of fatigue, heat stress, and crowd density, which increases mobility difficulties and fall risk, particularly among older adults and individuals with chronic diseases. Therefore, rapid operational mobility screening is required to identify risk and plan mobility. To support an operational mobility-classification workflow in a pre-Hajj setting, this study evaluated whether Timed Up and Go (TUG)-based stratification, combined with spatiotemporal gait and plantar pressure measurements, differentiates fall-risk categories. Methods: We conducted a cross-sectional study at a seasonal medical center near Al-Haram in Madinah Al-Munawwarah (21 May–3 June 2025) within the “I Lean On It” screening initiative. Participants completed the TUG and dynamic baropodometric gait assessments. We stratified the risk of falling as low (≤10 s), moderate (10.1–13.5 s), and high (>13.5 s) according to the TUG performance. We performed between-group comparisons using the Kruskal–Wallis test and evaluated the associations using Spearman’s correlation analysis. Results: Participants were classified as having low (n = 103), moderate (n = 24), or high (n = 29) fall risk. TUG performance significantly increased across the fall-risk groups. Significant between-group differences were observed in cadence, half-step length, walking speed, test duration, and functional mobility, whereas plantar pressure magnitude and gait symmetry did not differ significantly. Spearman correlation analysis showed significant negative correlations between TUG time and sex (rs = −0.357), half-step length (rs = −0.617), walking speed (rs = −0.577), and cadence (rs = −0.420). Significant positive correlations were observed with weight-bearing time (right: rs = 0.584; left: rs = 0.461), test duration (rs = 0.376), and number of steps acquired (rs = 0.356) (all p ≤ 0.003). Overall, TUG performance was primarily associated with dynamic gait and functional mobility. Conclusions: Integrated functional mobility and spatiotemporal gait screening significantly differentiate fall risk and provide clinically actionable mobility-support guidance in a mass-gathering pre-Hajj clinical workflow. Full article

Urban sustainable mobility requires understanding how people travel, which modes they use, and what impacts these choices generate. This study proposes a smart mobility analytics framework that integrates GPS traces, dynamic traffic variables, and machine learning to infer transport modes and sustainability metrics in Cuenca, Ecuador. Geospatial and kinematic data were collected at 1 Hz from 50 participants over four working weeks, yielding 8.99 million samples across five modes: walking, cycling, tram, bus, and private vehicles. A compact subset of physical and spatial predictors, derived from speed, acceleration, jerk, longitudinal forces, and distance to public transport routes, was selected using the Football Optimization Algorithm. A classification tree trained with a 70/15/15 train–validation–test split achieved an overall accuracy of 84.2%, with class precisions of about 99% for pedestrian and bicycle, 93% for tram, 76% for private vehicles, and 64% for bus. The classified trajectories show that walking and cycling account for approximately 65% of total travel time but only 2% of total distance and 1.7% of CO₂ emissions, whereas motorized modes generate more than 98% of emissions. Buses contribute nearly four times more CO₂ than private vehicles, despite carrying a larger passenger volume. The proposed framework delivers detailed, policy-relevant indicators to support low-carbon urban transport strategies. Full article

Skeleton-based gait sex classification can reduce reliance on appearance cues, yet its robustness under everyday walking disturbances remains under-quantified. Using PsyMo 2D pose sequences (90° side view), we render Common Objects in Context (COCO) keypoints into compact grayscale skeleton images, segment sequences into fixed-length 15-frame clips, and classify them with a 3D residual convolutional neural network (CNN) under a subject-wise split shared across four aggregated conditions: overall (A), occlusion/carrying disturbance (B), speed variation (C), and smartphone use (D). To avoid an arbitrary decision rule, we select a global operating threshold on the validation set by sweeping τ to maximize macro-F1, apply it unchanged to the held-out test set, and report a threshold-sensitivity check. Robustness is audited via condition-wise confusion matrices, subgroup precision/recall with 95% subject-level bootstrap confidence intervals, and subject-level probability overlap. To contextualize condition-dependent behavior, we quantify joint-group attribution shifts using Gradient-weighted Class Activation Mapping (Grad-CAM) and examine a coarse arm-swing proxy under smartphone use. Subject-level test accuracy ranged from 0.761 to 0.870 across conditions A–D, with uncertainty summarized by 95% subject-level bootstrap confidence intervals; performance was lowest in B, with increased male→female errors. Overall, these results provide a transparent audit-and-interpretation framework for assessing skeleton-based gait sex classification under realistic walking perturbations in practical evaluation scenarios. Full article

To address the limitations of conventional social force models in simulating high-density pedestrian crowds, this study proposes an enhanced model that incorporates visual perception constraints, group-type labeling, and collective avoidance mechanisms. Pedestrian trajectories were extracted from a bidirectional commercial street scenario using OpenCV, with YOLOv8 and DeepSORT employed for multiple object tracking. Analysis of pedestrian grouping patterns revealed that 52% of pedestrians walked in pairs, with distinct avoidance behaviors observed. The improved model integrates three key mechanisms: a restricted 120° forward visual field, group-type classification based on social relationships, and an exponentially formulated inter-group repulsive force. Simulation results in MATLAB R2023b demonstrate that the proposed model outperforms conventional approaches in multiple aspects: speed distribution (error < 8%); spatial density overlap (>85%); trajectory similarity (reduction of 32% in Dynamic Time Warping distance); and avoidance behavior accuracy (82% simulated vs. 85% measured). This model serves as a quantitative simulation tool and decision-making basis for the planning of pedestrian spaces, crowd organization management, and the optimization of emergency evacuation schemes in high-density pedestrian areas such as commercial streets and subway stations. Consequently, it contributes to enhancing pedestrian mobility efficiency and public safety, thereby supporting the development of a sustainable urban slow transportation system. Full article

Post-stroke gait dysfunction is biomechanically heterogeneous, yet biomechanically informed classifications of functional walking remain underdeveloped. In particular, there is a lack of clinically accessible methods for classifying gait deficits that account for propulsion impairments—a historically laboratory-dependent gait parameter requiring measurement with force plate systems. This study examined whether propulsion impairment can be classified by combining a global measure of walking function (i.e., the 10 m walk test speed) with specific measures of dynamic walking ability derived from the Functional Gait Assessment (FGA). Forty participants >6 months post-stroke completed biomechanical evaluations quantifying propulsion during walking and clinical assessments including the FGA. Multivariable stepwise regression identified the FGA items most strongly associated with paretic propulsion. Models augmented with these FGA items explained 15% greater variance in the paretic propulsion peak and 7% greater variance in paretic propulsion impulse compared with models using Comfortable Walking Speed (CWS) alone. Incorporating FGA items also yielded the highest overall accuracy (72.5% vs. 60% with CWS alone) and best per-class performance in propulsion severity classification. These findings establish the co-assessment of walking speed and targeted FGA items as a clinically feasible approach to biomechanically informed classification of post-stroke gait dysfunction. Full article

People with Parkinson’s disease (PD) experience mobility impairments that impact daily functioning, yet conventional clinical assessments provide limited insight into real-world mobility. This study evaluated motor-state classification and the concurrent validity of mobility metrics derived from augmented-reality (AR) glasses against a markerless motion capture system (Theia3D) during gamified AR exercises. Fifteen participants with PD completed five gamified AR exercises measured with both systems. Motor-state segments included straight walking, turning, squatting, and sit-to-stand/stand-to-sit transfers, from which the following mobility metrics were derived: step length, gait speed, cadence, transfer and squat durations, squat depth, turn duration, and peak turn angular velocity. We found excellent between-systems consistency for head position (X, Y, Z) and yaw-angle time series (ICC_(c,1) > 0.932). The AR-based motor-state classification showed high accuracy, with F1-scores of 0.947–1.000. Absolute agreement with Theia3D was excellent for all mobility metrics (ICC_(A,1) > 0.904), except for cadence during straight walking and peak angular velocity during turns, which were good and moderate (ICC_(A,1) = 0.890, ICC_(A,1) = 0.477, respectively). These results indicate that motor states and associated mobility metrics can be accurately derived during gamified AR exercises, verified in a controlled laboratory environment in people with mild to moderate PD, a necessary first step towards unobtrusive derivation of mobility metrics during in-clinic and at-home AR neurorehabilitation exercise programs. Full article

Wearable sensors have become valuable tools for assessing gait in both laboratory and free-living environments. However, detection of walking in free-living environments remains challenging, especially in clinical populations. Machine learning models may offer more robust gait identification, but most are trained on healthy participants, limiting their generalizability to other populations. To extend a previously validated machine learning model, an updated model was trained using an open dataset (PAMAP2), before progressively including training datasets with additional healthy participants and a clinical osteoarthritis population. The performance of the model in identifying walking was also evaluated using a frequency-based gait detection algorithm. The results showed that the model trained with all three datasets performed best in terms of activity classification, ultimately achieving a high accuracy of 96% on held-out test data. The model generally performed on par with the heuristic, frequency-based method for walking bout identification. However, for patients with slower gait speeds (<0.8 m/s), the machine learning model maintained high recall (>0.89), while the heuristic method performed poorly, with recall as low as 0.38. This study demonstrates the enhancement of existing model architectures by training with diverse datasets, highlighting the importance of dataset diversity when developing more robust models for clinical applications. Full article

Classifying human locomotor activities from wearable sensor data is an important high-level component of control schemes for many wearable robotic exoskeletons. In this study, we evaluated three machine learning models for classifying activity type (walking, running, jumping), speed, and surface incline using input data from body-worn inertial measurement units (IMUs) and e-textile insole pressure sensors. The IMUs were positioned on segments of the lower limb and pelvis during lab-based data collection from 16 healthy participants (11 men, 5 women), who walked and ran on a treadmill at a range of preset speeds and inclines. Logistic Regression (LR), Random Forest (RF), and Light Gradient-Boosting Machine (LGBM) models were trained, tuned, and scored on a validation data set (n = 14), and then evaluated on a test set (n = 2). The LGBM model consistently outperformed the other two, predicting activity and speed well, but not incline. Further analysis showed that LGBM performed equally well with data from a limited number of IMUs, and that speed prediction was challenged by inclusion of abnormally fast walking and slow running trials. Gyroscope data was most important to model performance. Overall, LGBM models show promise for implementing locomotor activity prediction from lower-limb-mounted IMU data recorded at different anatomical locations. Full article

The use of gait analysis to differentiate among paediatric populations with neurological and developmental conditions such as idiopathic toe walking (ITW), cerebral palsy (CP), and hereditary spastic paraplegia (HSP) remains challenging due to the insufficient precision of current diagnostic approaches, leading in some cases to misdiagnosis. Existing methods often isolate the analysis of gait variables, overlooking the whole complexity of biomechanical patterns and variations in motor control strategies. While previous studies have explored the use of statistical physics principles for the analysis of impaired gait patterns, gaps remain in integrating both kinematic and kinetic information or benchmarking these approaches against Deep Learning models. This study evaluates the robustness of statistical physics metrics in differentiating between normal and abnormal gait patterns and quantifies how the data source affects model performance. The analysis was conducted using gait data sets from two research institutions in Madrid and Dublin, with a total of 81 children with ITW, 300 with CP, 20 with HSP, and 127 typically developing children as controls. From each kinematic and kinetic time series, Shannon’s entropy, permutation entropy, weighted permutation entropy, and time irreversibility metrics were derived and used with Random Forest models. The classification accuracy of these features was compared to a ResNet Deep Learning model. Further analyses explored the effects of inter-laboratory comparisons and the spatiotemporal resolution of time series on classification performance and evaluated the impact of age and walking speed with linear mixed models. The results revealed that statistical physics metrics were able to differentiate among impaired gait patterns, achieving classification scores comparable to ResNet. The effects of walking speed and age on gait predictability and temporal organisation were observed as disease-specific patterns. However, performance differences across laboratories limit the generalisation of the trained models. These findings highlight the value of statistical physics metrics in the classification of children with different toe walking conditions and point towards the need of multimetric integration to improve diagnostic accuracy and gain a more comprehensive understanding of gait disorders. Full article

Accurate information on trail difficulty is essential for ensuring safety and enhancing the effectiveness of forest-based health and recreational activities. This study examined the physiological responses of middle-aged adults to varying trail difficulty levels across both controlled indoor and natural outdoor walking environments. A total of ten healthy individuals aged 40–50 years participated in walking tasks across three designated trail difficulty levels: Moderate, Difficult, and Very Difficult. Physiological indicators assessed included step speed (SS), step count (SC), rate of perceived exertion (RPE), heart rate (HR), oxygen saturation (OS), energy expenditure (EE), metabolic equivalents (MET), and oxygen consumption (VO₂). As trail difficulty increased, HR, RPE, VO₂, EE, and MET consistently showed upward trends, whereas SS and SC demonstrated significant decreases. Additionally, the outdoor setting imposed generally greater physiological demands compared to the indoor condition, suggesting that terrain complexity and elevation changes amplify physical exertion during real-world trail use. The findings contribute valuable empirical evidence for the design of individualized exercise programs, improved trail difficulty classifications, and the advancement of forest-based health promotion policies. Full article

Broiler gait score (GS) evaluation relies on manual assessments by experts, which can be laborious, hindering timely welfare management. Deep learning (DL) models, conversely, may serve as a cost-effective solution in evaluating GS via automated detection of broiler mobility. This study aimed to develop a vision-based YOLOv8 model to detect the locations of individual broilers, allowing for continuous tracking of birds within a pen and determining bird walking distances, speeds, idleness and movement ratios, and time at the feeder and drinker ratios. Then, Machine Learning (ML) models were developed to estimate the GS level from the mobility indicators in a lab setting. Ten broilers were color-coded and recorded via a top-view camera for 41 days. Their GS were assessed manually twice per week. The YOLOv8 model was trained, validated, and tested with 600, 150, and 50 images, respectively, and subsequently applied to the dataset yielding each broiler’s mobility indicators. The GS levels and mobility indicators were correlated through Ordinal Logistics (OL), Random Forest (RF), and Support Vector Machine (SVM) ML models. The YOLOv8 model was developed with 91% training, 89% testing, and 87% validation mean average precision (mAP) accuracies in identifying color-coded broilers. After tracking, the model estimated an average of 472.26 ± 234.18 cm hourly distance traveled and 0.13 ± 0.07 cm/s speed by a broiler. It was found that with deteriorated GS levels (i.e., worse walking ability), broilers walked shorter distances (p = 0.001), had lower speeds (p = 0.001), were increasingly idle and less mobile, and were increasingly stationed near or around the feeder. The movement ratio, average hourly walking distance, hourly average speed, and age variables were found to be the most significant variables (p < 0.005) in predicting GS levels. These variables were further reduced to one, the average hourly walking distance, because of high collinearity and were used to predict the GS with ML models. The RF model, outperforming others, was able to predict GS with a generalized R² of 0.62, root mean squared error (RMSE) of 0.54, and 65% classification accuracy. Full article

Background: Using the local divergence exponent (LDE), it has been concluded that walking stability is impaired in people with multiple sclerosis (pwMS). However, the use of several calculation approaches hinders comparisons across studies. We aimed to determine whether using different parameters for state space reconstruction to calculate LDE affects the classification of pwMS. Methods: A total of 55 pwMS and 23 controls walked up and down a 20 m corridor for 5 min. The LDE was calculated using three different combinations of n-dimensions (d_E) and time delays (τ): (a) trial-specific, (b) median across subjects, and (c) fixed d_E = 5 and τ = 10. The LDE was calculated using vertical (VT), mediolateral (ML), and anteroposterior (AP) accelerations, the norm (N), and 3D data from sensors placed on the sternum and lumbar. Classification accuracy across results obtained with different parameter combinations was compared using a Quadratic Discriminant Analysis (QDA). Results: The best classification accuracy, 84%, was achieved when using the LDE obtained with norm acceleration data from the sternum sensor with a fixed d_E = 5 and τ = 10 and considering speed as a covariate. Lumbar LDEs were less accurate than sternum LDEs. Conclusions: LDEs calculated with a fixed d_E = 5 and τ = 10 for the norm acceleration from a sternum-placed sensor can best classify pwMS. Using fixed parameters for the state space reconstruction, and consequently LDE calculation, can simplify the implementation of the LDE as a mobility biomarker in MS and provides evidence for future consensus for its calculation. Full article

This study analyzes factors influencing injury severity in motorcycle accidents involving non-helmeted riders using Bayesian spatiotemporal logistic models. Five models were developed, four of which incorporated different spatiotemporal configurations, including spatial, temporal, and spatiotemporal interaction error terms. The results indicate that the optimal model integrated Leroux CAR spatial priors, temporal random walks, and interaction terms, achieving 86.74% classification accuracy, with a 3% reduction in the DIC value; obtaining the lowest numerical fit demonstrating spatiotemporal interactions is critical for capturing complex risk patterns (e.g., rain amplifying nighttime collision severity). The results highlight rain (OR = 1.53), age ≥ 50 (OR = 1.90), and bi-directional roads (OR = 1.82) as critical risk factors. Based on these findings, several sustainable traffic safety strategies are proposed. Short-term measures include IoT-based dynamic speed control on high-risk roads and app-enforced helmet checks via ride-hailing platforms. Long-term strategies integrate age-specific behavioral training focusing on hazard perception and reaction time improvement, which reduced elderly fatalities by 18% in Japan’s “Silver Rider” program by directly modifying high-risk riding habits (non-helmets). These solutions, validated by global case studies, demonstrate that helmet use could mitigate over 60% of severe head injuries in these high-risk scenarios, promoting sustainable traffic governance through spatiotemporal risk targeting and helmet enforcement. Full article

This study classified cows’ foraging behaviors using machine learning (ML) models evaluated through random test split (RTS) and cross-validation (CV) data partition methods. Models included Perceptron, Logistic Regression, Support Vector Machine, K-Nearest Neighbors, Random Forest (RF), and XGBoost (XGB). These models classified activity states (active vs. static), foraging behaviors (grazing (GR), resting (RE), walking (W), ruminating (RU)), posture states (standing up (SU) vs. lying down (LD)), and posture combinations with rumination and resting behaviors (RU_SU, RU_LD, RE_SU, and RE_LD). XGB achieved the highest accuracy for state classification (74.5% RTS, 74.2% CV) and foraging behavior (69.4% CV). RF outperformed XGB in other classifications, including GR, RE, and RU (62.9% CV vs. 56.4% RTS), posture (83.9% CV vs. 79.4% RTS), and behaviors-by-posture (58.8% CV vs. 56.4% RTS). Key predictors varied: speed and Actindex were crucial for GR and W when increasing and for RE and RU when decreasing. X low values were linked to RE_SU and RU_SU, while X and Z influenced RE_LD more. RTS showed higher accuracy in activity states classification while CV in foraging behaviors and by posture classification. These results emphasize CV in RF’s reliability in managing complex behavioral patterns and the importance of continuous recording devices and movement data to monitor cattle behavior accurately. Full article

Background: Indirect calorimetry is the gold standard field-testing technique for measuring energy expenditure and exercise intensity based on the volume of oxygen consumed (VO₂, mL O₂/min). Although heart rate is often used as a proxy for VO₂, heart rate-based estimates of VO₂ may be inaccurate after stroke due to changes in the heart rate–VO₂ relationship. Our objective was to evaluate in people post stroke the accuracy of using heart rate to estimate relative walking VO₂ (wVO₂) and classify exercise intensity. Moreover, we sought to determine if estimation accuracy could be improved by including clinical variables related to patients’ function and health in the estimation. Methods: Sixteen individuals post stroke completed treadmill walking exercises with concurrent indirect calorimetry and heart rate monitoring. Using 70% of the data, forward selection regression with repeated k-fold cross-validation was used to build wVO₂ estimation equations that use heart rate alone and together with clinical variables available at the point-of-care (i.e., BMI, age, sex, and comfortable walking speed). The remaining 30% of the data were used to evaluate accuracy by comparing (1) the estimated and actual wVO₂ measurements and (2) the exercise intensity classifications based on metabolic equivalents (METs) calculated using the estimated and actual wVO₂ measurements. Results: Heart rate-based wVO₂ estimates were inaccurate (MAE = 3.11 mL O₂/kg/min) and unreliable (ICC = 0.68). Incorporating BMI, age, and sex in the estimation resulted in improvements in accuracy (MAE Δ: −36.01%, MAE = 1.99 mL O₂/kg/min) and reliability (ICC Δ: +20, ICC = 0.88). Improved exercise intensity classifications were also observed, with higher accuracy (Δ: +29.85%, from 0.67 to 0.87), kappa (Δ: +108.33%, from 0.36 to 0.75), sensitivity (Δ: +30.43%, from 0.46 to 0.60), and specificity (Δ: +17.95%, from 0.78 to 0.92). Conclusions: In people post stroke, heart rate-based wVO₂ estimations are inaccurate but can be substantially improved by incorporating clinical variables readily available at the point of care. Full article