Search Results (31)

Differentiating genuine hemiplegic gait (HG) in stroke survivors from hemiplegic-like gait voluntarily imitated by healthy adults (MHG) is essential for reliable assessment and intervention planning. Treadmill-based gait data were obtained from 79 participants—39 stroke patients (HG) and 40 healthy adults—instructed to mimic HG (MHG). Forty-eight spatiotemporal and force-related variables were extracted. Random Forest, support vector machine (SVM), and logistic regression classifiers were trained with (i) the full feature set and (ii) the 10 most important features selected via Random Forest Gini importance. Performance was assessed with 5-fold stratified cross-validation and an 80/20 hold-out test, using accuracy, F1-score, and the area under the receiver operating characteristic curve (AUC). All models achieved high discrimination (AUC > 0.93). The SVM attained perfect discrimination (AUC = 1.000, test set) with the full feature set and maintained excellent accuracy (AUC = 0.983) with only the top 10 features. Temporal asymmetries, delayed vertical ground reaction force peaks, and mediolateral spatial instability ranked highest in importance. Reduced-feature models showed negligible performance loss, highlighting their parsimony and interpretability. Supervised machine learning algorithms can accurately distinguish true hemiplegic gait from mimicked patterns using a compact subset of gait features. The findings support data-driven, time-efficient gait assessments for clinical neurorehabilitation and for validating experimental protocols that rely on gait imitation. Full article

Curved non-motorized treadmills (cNMTs) have been demonstrated to reduce impact accelerations in comparison with motorized treadmills (MTs). Most studies have analyzed impacts in the time domain, but analysis in the frequency domain can provide useful information associated with the increase in the running risk of injury. The purpose of this study was to analyze the frequency components (low- and high-frequency bands) of impact accelerations, countermovement jump (CMJ) height, and perceived comfort during a prolonged run on different surfaces: MT, cNMT, and overground (OVG). Twenty-one recreational runners completed three randomized prolonged running tests on cNMT, MT, and OVG for 30 min (80% of the individual maximal aerobic speed). Impact accelerations were registered at minutes 5 and 30 of the test, the countermovement jump test (CMJ) was performed before (PreTest) and after (PostTest) the test, and perceived comfort was determined at the end of each test. A two-way repeated-measures analysis of variance (significance at p < 0.05) showed a reduction on cNMT in both low- and high-frequency bands of impact accelerations, such as head power (p < 0.001, ESd = 3.0) on the cNMT vs. the MT and tibia peak power (p = 0.001, ESd = 2.2) on the cNMT vs. OVG. However, cNMT was perceived as the least comfortable surface by runners. The prolonged running effect decreased impact accelerations during the treadmill running test (MT and cNMT) in the low-frequency band, while CMJ height decreased (p = 0.024, ESd = 1.4) during the PostTest vs. PreTest with the cNMT. Using a cNMT could be an interesting strategy for load reduction in long-distance runners or in return-to-play rehabilitation protocols. Full article

Vertical ground reaction force (vGRF) plays an important role in the study of running-related injuries (RRIs). This study explores the synchronization method between inertial measurement unit (IMU) and vGRF data of running and develops ANN models to accurately predict vGRF. Fifteen runners participated in this study. Acceleration data and vGRF values of eight rearfoot strikers and seven forefoot strikers running at 12, 14, and 16 km/h were collected by a single IMU and an instrumented treadmill. The sliding time window synchronization (STWS) algorithm was developed to sync IMU data with vGRF data. The wavelet neural network model (WNN) and feed-forward neural network model (FFNN) were adapted to predict vGRF using three-axis or sagittal-axis acceleration data in the stance phase, respectively. One rearfoot striker and one forefoot striker were randomly selected as a test set, while the other participants formed training sets. After synchronization, mean absolute errors for stride time of the IMU and vGRF data were less than 11.2 ms. The coefficient of multiple correlations for vGRF measured curves and predicted curves was more than 0.97. The normalized root mean square errors (NRMSEs) between two curves were 4.6~9.2%, and R² was 0.93~0.99. For peak vGRF, the NRMSEs were 1.6~8.2%, except for rearfoot strike runners at 16 km/h using the FFNN model (10.7% and 11.1%). The Bland–Altman plots indicate that the errors for both the WNN and FFNN models are within acceptable limits. The STWS algorithm can effectively achieve the data synchronization between the IMU and the force plate during running. Both WNN and FFNN models demonstrated good accuracy and agreement in predicting vGRF. Using sagittal-axis acceleration data may be an ideal model with good prediction accuracy and less input data. This work provides direction for developing ANN models of personalized monitoring of lower limb load. Full article

Smartwatches are one of the most relevant fitness trends of the past two decades, and they collect increasing amounts of health and movement data. The accuracy of these data may be questionable and requires further investigation. Therefore, the aim of the present study is to validate smartwatches for use in triathlon training. Ten different smartwatches were tested for accuracy in measuring heart rates, distances (via global navigation satellite systems, GNSSs), swim stroke rates and the number of swim laps in a 50 m Olympic-size pool. The optical heart rate measurement function of each smartwatch was compared to that of a chest strap. Thirty participants (15 females, 15 males) ran five 3 min intervals on a motorised treadmill to evaluate the accuracy of the heart rate measurements. Moreover, for each smartwatch, running and cycling distance tracking was tested over six runs of 4000 m on a 400 m tartan stadium track, six hilly outdoor runs over 3.4 km, and four repetitions of a 36.8 km road bike course, respectively. Three swimming protocols ranging from 200 m to 400 m were performed in triplicate in a 50 m Olympic-size pool, evaluating the tracked distance and the detected number of strokes. The mean absolute percentage errors (MAPEs) for the average heart rate measurements varied between 3.1% and 8.3%, with the coefficient of determination ranging from 0.22 to 0.79. MAPE results ranged from 0.8% to 12.1% for the 4000 m run on the 400 m track, from 0.2% to 7.5% for the 3.4 km outdoor run, and from 0.0% to 4.2% for the 36.8 km bike ride. For the swimming tests, in contrast, the deviations from the true distance varied greatly, starting at a 0.0% MAPE for the 400 m freestyle and reaching 91.7% for the 200 m medley with style changes every 25 m. In summary, for some of the smartwatches, the measurement results deviated substantially from the true values. Measurements taken while road cycling over longer distances with only a few curves were in relative terms more accurate than those taken during outdoor runs and even more accurate than those taken on the 400 m track. In the swimming exercises, the accuracy of the measured distances was severely deteriorated by the medley changes among the majority of the smartwatches. Altogether, the results of this study should help in assessing the accuracy and thus the suitability of smartwatches for general triathlon training. Full article

Inertial measurement units (IMUs) have been validated for measuring sagittal plane lower-limb kinematics during moderate-speed running, but their accuracy at maximal speeds remains less understood. This study aimed to assess IMU measurement accuracy during high-speed running and maximal effort sprinting on a curved non-motorized treadmill using discrete (Bland–Altman analysis) and continuous (root mean square error [RMSE], normalised RMSE, Pearson correlation, and statistical parametric mapping analysis [SPM]) metrics. The hip, knee, and ankle flexions and the pelvic orientation (tilt, obliquity, and rotation) were captured concurrently from both IMU and optical motion capture systems, as 20 participants ran steadily at 70%, 80%, 90%, and 100% of their maximal effort sprinting speed (5.36 ± 0.55, 6.02 ± 0.60, 6.66 ± 0.71, and 7.09 ± 0.73 m/s, respectively). Bland–Altman analysis indicated a systematic bias within ±1° for the peak pelvic tilt, rotation, and lower-limb kinematics and −3.3° to −4.1° for the pelvic obliquity. The SPM analysis demonstrated a good agreement in the hip and knee flexion angles for most phases of the stride cycle, albeit with significant differences noted around the ipsilateral toe-off. The RMSE ranged from 4.3° (pelvic obliquity at 70% speed) to 7.8° (hip flexion at 100% speed). Correlation coefficients ranged from 0.44 (pelvic tilt at 90%) to 0.99 (hip and knee flexions at all speeds). Running speed minimally but significantly affected the RMSE for the hip and ankle flexions. The present IMU system is effective for measuring lower-limb kinematics during sprinting, but the pelvic orientation estimation was less accurate. Full article

In this paper, we present a method of gait analysis on a treadmill based on pressure distribution. We aimed to model the gait patterns of a subject walking at a constant speed on a treadmill based on differences in current consumption. The changes in current consumption were converted into pressure distribution curves, and then specific features were extracted. The extracted features were used to model the walking pattern on a treadmill. To verify the validity of our proposed feature-based gait pattern modeling, we conducted experiments by gender, age, BMI (body mass index), and step-to-step symmetry. The experimental results showed that the heavier the subject, the higher the value of each feature. In particular, our feature point-based gait modeling provides an index that can help determine whether a subject’s gait is abnormal, depending on the difference between the features. Full article

Machine learning-based gait systems facilitate the real-time control of gait assistive technologies in neurological conditions. Improving such systems needs the identification of kinematic signals from inertial measurement unit wearables (IMUs) that are robust across different walking conditions without extensive data processing. We quantify changes in two kinematic signals, acceleration and angular velocity, from IMUs worn on the frontal plane of bilateral shanks and thighs in 30 adolescents (8–18 years) on a treadmills and outdoor overground walking at three different speeds (self-selected, slow, and fast). Primary curve-based analyses included similarity analyses such as cosine, Euclidean distance, Poincare analysis, and a newly defined bilateral symmetry dissimilarity test (BSDT). Analysis indicated that superior–inferior shank acceleration (SI shank Acc) and medial–lateral shank angular velocity (ML shank AV) demonstrated no differences to the control signal in BSDT, indicating the least variability across the different walking conditions. Both SI shank Acc and ML shank AV were also robust in Poincare analysis. Secondary parameter-based similarity analyses with conventional spatiotemporal gait parameters were also performed. This normative dataset of walking reports raw signal kinematics that demonstrate the least to most variability in switching between treadmill and outdoor walking to help guide future machine learning models to assist gait in pediatric neurological conditions. Full article

This study’s aim was to assess FreeStyle Libre Flash glucose monitoring (FGM) performance during an oral glucose tolerance test (OGTT) and treadmill exercise in healthy adolescents. This should advance the feasibility and utility of user-friendly technologies for metabolic assessments in adolescents. Seventeen healthy adolescents (nine girls aged 12.8 ± 0.9 years) performed an OGTT and submaximal and maximal treadmill exercise tests in a laboratory setting. The scanned interstitial fluid glucose concentration ([ISFG]) obtained by FGM was compared against finger-prick capillary plasma glucose concentration ([CPG]) at 0 (pre-OGTT), −15, −30, −60, −120 min post-OGTT, pre-, mid-, post- submaximal exercise, and pre- and post- maximal exercise. Overall mean absolute relative difference (MARD) was 13.1 ± 8.5%, and 68% (n = 113) of the paired glucose data met the ISO 15197:2013 criteria. For clinical accuracy, 84% and 16% of FGM readings were within zones A and B in the Consensus Error Grid (CEG), respectively, which met the ISO 15197:2013 criteria of having at least 99% of results within these zones. Scanned [ISFG] were statistically lower than [CPG] at 15 (−1.16 mmol∙L⁻¹, p < 0.001) and 30 min (−0.74 mmol∙L⁻¹, p = 0.041) post-OGTT. Yet, post-OGTT glycaemic responses assessed by total and incremental areas under the curve (AUCs) were not significantly different, with trivial to small effect sizes (p ≥ 0.084, d = 0.14–0.45). Further, [ISFGs] were not different from [CPGs] during submaximal and maximal exercise tests (interaction p ≥ 0.614). FGM can be a feasible alternative to reflect postprandial glycaemia (AUCs) in healthy adolescents who may not endure repeated finger pricks. Full article

Hunger and satiety are controlled by several physiological mechanisms, including pancreatic and gastrointestinal hormones. While the influence of exercise and fasting have been described individually, in relation to these hormones, there is a paucity of work showing the effects of the two modalities (fasting and exercise) combined. Twenty healthy adults (11 males, 9 females) completed both conditions of this study, each consisting of a 36-h water-only fast. One of the fasts began with treadmill exercise, and the differences between the conditions on various appetite hormones were measured every 12 h. The difference in the area under the curve between conditions for ghrelin was 211.8 ± 73.1 pg/mL (F = 8.40, p < 0.0105), and, for GLP-1, it was −1867.9 ± 850.4 pg/mL (F = 4.82, p < 0.0422). No significant differences were noted for areas under the curve between conditions for leptin, PP, PYY, insulin, or GIP. Initiating a fast with exercise lowers ghrelin concentrations and elevates GLP-1 concentrations. Given that ghrelin elicits feelings of hunger and GLP-1 signals feelings of satiety, adding exercise to the beginning of a fast may reduce some of the biological drive of hunger, which could make fasting more tolerable, leading to better adherence and more significant health outcomes. Full article

Iliotibial band syndrome (ITBS) is a common running related injury. While previous studies have evaluated the relationship between biomechanical variables and ITBS, most have found limited evidence, particularly with measures related to ground reaction force (GRF). The purpose of this study was to use a classification and regression tree (CART) analysis to determine whether the combination of GRF measures would be strongly associated in runners with ITBS. A cross-sectional study was performed at an outpatient center focused on running injuries. A convenience sample of 52 runners with ITBS, assessed between September 2012 and July 2022, were evaluated for eligibility, from which, 30 rearfoot strike runners with ITBS and no secondary running-related injuries were selected. Injured runners were matched to 30 healthy controls from a normative database. Each ran on an instrumented treadmill at a self-selected speed. GRF variables were calculated, including peak GRFs, loading rates, and impulses. CART analysis was performed to identify interactions between GRF data and runners with ITBS. An ROC curve was executed, to determine the accuracy of the model. Posterior GRF impulse (PGRFI), anterior GRF (AGRFI), peak anterior GRF (PAGRF), and vertical stiffness at initial loading (VSIL) all emerged as variables associated with ITBS in the CART analysis. The model was able to correctly identify 25 (83.3%) runners with ITBS and 25 (83.3%) controls. The area under the ROC curve (accuracy) was 0.87 (95% CI, 0.77–0.96; SE, 0.04; p < 0.001). In conclusion, interactions between GRF variables were associated with ITBS in runners. The best classification included interactions between PGRFI, AGRFI, AGRFP, and VSIL, using specific cut-off values. Loading rates were not independently associated with ITBS. Full article

Several methods could be used to measure the forces from skis or roller skis in cross-country skiing. Equipment that could measure medio-lateral forces may be of good help for investigating the relevant skating techniques. The aim of this study was to validate a pair of newly designed two-dimensional force measurement roller skis. The vertical and medio-lateral forces which were perpendicular to the body of the roller ski could be measured. Forces were resolved into the global coordinate system and compared with the force components measured by a force plate. A static and dynamic loading situation for the force measurement roller ski was performed to reveal the validity of the system. To demonstrate whether the force measurement roller ski would affect roller skiing performance on a treadmill, a maximum speed test with the V2 technique was performed by using both normal and force measurement roller skis. The force-time curves obtained by these two different force measurement systems were shown to have high similarity (coefficient of multiple correlations > 0.940). The absolute difference for the forces in the X and Z directions over one push-off cycle was 3.9–33.3 N. The extra weight (333 g) of the force measurement roller ski did not affect the performance of the skiers. Overall, the newly designed two-dimensional force measurement roller ski in this study is valid for use in future research during daily training for skate skiing techniques. Full article

Accelerometers can estimate the intensity, frequency, and duration of physical activity in healthy adults. Although thresholds to distinguish varying levels of activity intensity using the Actigraph wGT3X-B have been established for the general population, their accuracy for Huntington’s disease (HD) is unknown. We aimed to define and cross-validate accelerometer cut-points for different walking speeds in adults with mild to moderate HD. A cross-sectional, multicentre, case-control, observational study was conducted with a convenience sample of 13 symptomatic ambulatory HD participants. The accelerometer was placed around the right hip, and a heart monitor was fitted around the chest to monitor heart rate variability. Participants walked on a treadmill at three speeds with light, moderate and vigorous intensities. Correlation and receiver operation curve analyses were performed between the accelerometer magnitude vector with relative oxygen and heart rate. Optimal cut-points for walking speeds of 3.2 km/h were ≤2852; 5.2 km/h: >2852 to ≤4117, and in increments until their maximum velocity: >4117. Our results support the application of the disease-specific cut-points for quantifying physical activity in patients with mild to moderate HD and promoting healthy lifestyle interventions. Full article

Coronary artery disease (CAD) is one of the most common causes of heart ailments; many patients with CAD do not exhibit initial symptoms. An electrocardiogram (ECG) is a diagnostic tool widely used to capture the abnormal activity of the heart and help with diagnoses. Assessing ECG signals may be challenging and time-consuming. Identifying abnormal ECG morphologies, especially in low amplitude curves, may be prone to error. Hence, a system that can automatically detect and assess the ECG and treadmill test ECG (TMT-ECG) signals will be helpful to the medical industry in detecting CAD. In the present work, we developed an intelligent system that can predict CAD, based on ECG and TMT signals more accurately than any other system developed thus far. The distinct convolutional neural network (CNN) architecture deals with single-lead and multi-lead (12-lead) ECG and TMT-ECG data effectively. While most artificial intelligence-based systems rely on the universal dataset, the current work used clinical lab data collected from a renowned hospital in the neighborhood. ECG and TMT-ECG graphs of normal and CAD patients were collected in the form of scanned reports. One-dimensional ECG data with all possible features were extracted from the scanned report with the help of a modified image processing method. This feature extraction procedure was integrated with the optimized architecture of the CNN model leading to a novel prediction system for CAD. The automated computer-assisted system helps in the detection and medication of CAD with a high prediction accuracy of 99%. Full article

This study evaluated two approaches for estimating the total propulsive force on a skier’s center of mass (COM) with double-poling (DP) and V2-skating (V2) skiing techniques. We also assessed the accuracy and the stability of each approach by changing the speed and the incline of the treadmill. A total of 10 cross-country skiers participated in this study. Force measurement bindings, pole force sensors, and an eight-camera Vicon system were used for data collection. The coefficient of multiple correlation (CMC) was calculated to evaluate the similarity between the force curves. Mean absolute force differences between the estimated values and the reference value were computed to evaluate the accuracy of each approach. In both DP and V2 techniques, the force–time curves of the forward component of the translational force were similar to the reference value (CMC: 0.832–0.936). The similarity between the force and time curves of the forward component of the ground reaction force (GRF) and the reference value was, however, greater (CMC: 0.879–0.955). Both approaches can estimate the trend of the force–time curve of the propulsive force properly. An approach by calculating the forward component of GRF is a more appropriate method due to a better accuracy. Full article

Muscle synergy theory assumes that the central nervous system generates a wide range of complex motor outputs by recruiting muscle synergies with different strengths and timings. The current understanding is that a common set of muscle synergies underlies unperturbed as well as perturbed walking at self-selected speeds. However, it is not known whether this is the case for substantially slower walking. The aim of this study was to investigate whether a shared set of muscle synergies underlies balance recovery responses following inward- and outward-directed perturbations in the mediolateral direction at various perturbation onsets and walking speeds. Twelve healthy subjects walked at three walking speeds (0.4, 0.6, and 0.8 m/s) on a treadmill while perturbations were applied to the pelvis using the balance assessment robot. A set of sixteen EMG signals, i.e., eight muscles per leg, was measured and decomposed into muscle synergies and weighting curves using non-negative matrix factorization. The muscles included were left and right tibialis anterior, soleus, gastrocnemius medialis, gastrocnemius lateralis, rectus femoris, hamstring, gluteus medius, and gluteus maximus. In general, four muscle synergies were needed to adequately reconstruct the data. Muscle synergies were similar for unperturbed and perturbed walking at a high walking speed (0.8 m/s). However, the number of similar muscle synergies between perturbed and unperturbed walking was significantly lower for low walking speeds (0.4 and 0.6 m/s). These results indicate that shared muscle synergies underlying perturbed and unperturbed walking are less present during slow walking compared to fast walking. Full article