Search Results (124)

Acoustic gait recognition is an emerging passive biometric modality that identifies individuals by unique walking sound patterns. This work presents a reproducible benchmarking framework for convolutional neural network (CNN)-based acoustic gait recognition, providing a systematic evaluation methodology across varying identity pool sizes. Raw footstep recordings from the AFPILD dataset were converted into 128-bin mel-spectrograms and used to train a compact CNN across identity pool sizes from 10 to 40 subjects. To ensure statistical reliability, a three-times-repeated five-fold stratified cross-validation protocol was implemented. Experimental results demonstrate strong discriminative capability, with validation accuracy reaching 94.92% and Equal Error Rate (EER) of 1.31% for the 40-subject configuration. A multi-seed subset validation experiment across five independent random subject draws per pool size confirmed that the observed scaling trend is consistent across subset compositions rather than an artifact of a single subject selection. Additional analysis confirmed the framework’s resilience to moderate environmental noise and its superiority over classical Mel-Frequency Cepstral Coefficients paired with a Support Vector Machine (MFCC-SVM) and Convolutional Recurrent Neural Network (CRNN) baselines, supporting the feasibility of acoustic gait recognition as a passive biometric modality. Full article

Sarcopenia is an age-related condition characterized by a decline in skeletal muscle mass and function, leading to impaired mobility and an increased risk of adverse health outcomes. However, the neuromuscular mechanisms underlying gait dysfunction in sarcopenia remain incompletely understood. In this study, individuals with sarcopenia and age-matched healthy controls were recruited. Gait parameters were assessed using a motion capture system and quantified through spatiotemporal analysis, muscle activity was evaluated using surface electromyography (sEMG) with phase-specific activation metrics, and cortical activity was measured using electroencephalography (EEG) and further analyzed using spectral analysis and partial directed coherence (PDC)-based graph-theoretical measures to assess frequency-specific functional connectivity. Individuals with sarcopenia exhibited significantly reduced gait speed and shorter step length, along with prolonged loading response and pre-swing phases. Among the recorded muscles, the tibialis anterior (TA) showed significant alterations, characterized by an increased and earlier first activation peak and a reduced and delayed second peak during the gait cycle. Phase-specific analysis revealed increased TA activity during the loading response phase and decreased activity during the pre-swing phase. EEG analysis revealed beta-band-specific alterations, with increased node strength and node degree in the frontal and central regions and elevated node strength in the parietal region, while no significant differences were observed in the delta, theta, alpha, or gamma bands. These findings suggest that sarcopenia is associated with neuromuscular alterations. The coexistence of increased beta-band functional connectivity strength and persistent gait impairment may reflect inefficient compensation, in which increased neural recruitment does not fully restore gait function. These results highlight the importance of targeting neuromuscular coordination in rehabilitation. Full article

(1) Background: Gait analysis technologies have advanced; however, traditional systems like optical motion capture are lab-bound and costly, limiting rehabilitation monitoring. This cross-sectional study evaluates self-powered triboelectric nanogenerator (TENG) insoles combined with IMU sensors to assess gait asymmetry, plantar pressure signatures, age effects and injury history in rehabilitation patients, aiming to enable portable, battery-free phenotyping. (2) Methods: Fifty-three patients (22 females, 31 males; age, 29 ± 26 years) from Astana clinics with trauma histories (e.g., spine, ankle, fractures) and 10 healthy references underwent a 2 min walk test (2MWT). TENG insoles captured plantar loading; ankle/knee IMUs measured spatiotemporal parameters (cadence, asymmetry). The data were normalized; the analyses used an ANOVA and correlations (Python 3.14.3). (3) Results: The TENG sensors showed force/frequency linearity (up to 10 V at 20 N). The cadence averaged 101 ± 10 steps/min, declining with age (r = −0.31, p = 0.03) and fractures (r = −0.23, p = 0.04). The asymmetry varied (−54% to +31%) without category differences. Flatfoot (55%) was linked to lateral loading shifts; condition-specific waveform signatures emerged (e.g., lateral heel in ankle issues). (4) TENG-IMU systems feasibly capture gait phenotypes in heterogeneous cohorts, supporting out-of-lab monitoring for personalized rehabilitation without batteries. Prospective validation is required for further practical implications. Full article

Smart insoles equipped with pressure sensor matrices are increasingly used for gait analysis, yet high-density arrays compromise battery life and data throughput. This study aims to identify the optimal sparse sensor layout required to accurately estimate the Center of Pressure (CoP) by analyzing the trade-off between sensor number, spatial placement, and reconstruction error. Plantar pressure data were collected from twelve healthy participants walking at a self-selected speed using 16-sensor connected insoles. A combinatorial algorithm evaluated all $2^{16}-1$ possible sensor combinations to minimize the Root Mean Square Error (RMSE) in the antero-posterior, medio-lateral, and global Euclidean directions. Results reveal a non-linear convergence of accuracy that depends on the spatial axis. For longitudinal and global progression, a clear inflection point achieving sub-centimetric accuracy (RMSE < 5 mm) is reached at seven sensors. In contrast, medio-lateral tracking shows its largest discrete error reduction at five sensors, followed by gradual improvements at higher densities. Anatomical frequency analysis highlights distinct spatial requirements: the posterior heel is consistently selected for medio-lateral accuracy, while the lateral arch and metatarsal regions are critical for longitudinal progression. These findings suggest that while a minimum of seven strategically placed sensors enables robust CoP tracking across all spatial axes, optimal hardware design should remain task-specific. This work provides a data-driven framework for the development of energy-efficient wearable gait monitoring systems. Full article

The evaluation of biomechanical parameters in four-beat gaited horses remains limited by the subjectiveness and complexity of current standard methods. Through a deep learning approach, we aimed to infer dissociation % using only acoustic signals. A total of 268 audio samples were extracted from publicly available videos featuring three Brazilian horse breeds (Mangalarga Marchador, Campolina, and Piquira) performing marcha batida and marcha picada. Acoustic features, including root mean square energy (RMS), zero-crossing rate (ZCR), and 13 Mel-frequency cepstral coefficients (MFCCs), were extracted and used to train a long short-term memory (LSTM) neural network. The model accurately predicted the time intervals between successive hoof–ground contacts (R² = 0.98; MAE = 0.0071), enabling the calculation of the dissociation %. While no significant differences were found between gait types and dissociation %, breed-related differences in both mean hoof–ground contact interval and dissociation were observed, with 8 acoustic features demonstrating discriminative power. Our results suggest that hoof–ground contact patterns can be quantified objectively from audio alone, offering a practical and non-invasive method for gait analysis. The approach holds potential for applications in breed standardization, selection, and digital locomotion phenotyping of horse populations. Full article

Background/Objectives: Multiple epiphyseal dysplasia (MED) is a clinically and genetically heterogeneous group of disorders characterized by a waddling gait, joint pain, and early-onset osteoarthritis. The aim of this study was to compare the genetic characteristics and long-term clinical follow-up findings of 22 patients with MED from 17 unrelated families. Methods: Molecular diagnosis was performed using clinical exome analysis and exome sequencing. Seventeen children were followed for a median of 5.5 years. Results: Eighteen disease-related variants were identified: 47% in COMP, 11.8% each in COL9A2 and COL9A3 in a monoallelic state, 17.6% in SLC26A2, and 11.8% each in MATN3 and CANT1 in a biallelic state. Some COMP mutations previously identified in pseudoachondroplasia, an allelic disorder of MED1, were shown in our study to exhibit a typical MED1 or intermediate phenotype. In contrast, it was confirmed that certain mutations in SLC26A2 lead to MED4 phenotype. Furthermore, it has been observed that biallelic variants in MATN3 may be associated with the MED5 phenotype. In patients with MED2 and MED3, the knee joint is affected, while in other types, the hip joint is predominantly affected. In 15 children followed until ages 11–18, height decreased slightly as they grew older but remained normal or at the lower limit, and slow progression was observed in the waddling gait and joint pain, except in the intermediate form. Conclusions: This study reveals the frequency of disease-related variants, including seven novel ones, in genes leading to MED1–5 and 7 phenotypes, and expands the spectrum of genetic and clinical phenotypes. Full article

In this paper, an improved state-space method (SSM) is proposed for gait feature extraction. By introducing zero-phase component analysis Whitening (ZCA Whitening) and an algorithm to search estimated echo as the preprocessing method, pedestrian echoes are divided into three groups according to the frequency probability density: torso, feet, and other segments. Two channels of echoes are selected as inputs to the SSM, which is employed to identify the corresponding micro-Doppler trajectory. On this basis, five gait features of torso amplitude, stride length, walking cycle, torso maximum speed, and feet maximum speed are extracted. Simulation based on the Boulic model, compared with the traditional SSM, demonstrated that there is no need to estimate the model order and that a more accurate torso micro-Doppler trajectory and effective micro-motion features of the feet can be obtained by the proposed method. Finally, 77 GHz FMCW radar was used to collect the echoes of four pedestrians. The classifier was designed based on a support vector machine (SVM), and the classification experiment verified the effectiveness of the extracted gait features. Full article

Background and Objectives: Parkinson’s disease (PD) is a neurodegenerative disorder marked by bradykinesia, rigidity, and tremor. While deep brain stimulation (DBS) of the subthalamic nucleus (STN) and globus pallidus internus (GPi) effectively alleviates motor symptoms, the potential of targeting the substantia nigra pars reticulata (SNr) is less understood. This study investigates the effects of mid-term DBS of the SNr on motor function and neuroplasticity in a 6-hydroxydopamine (6-OHDA) rat model of PD. Methods: Adult male Sprague-Dawley rats (280–300 g) were divided into healthy control (n = 10), PD (n = 9), sham-DBS (n = 7), and SNr-DBS (n = 7) groups. Bilateral striatal 6-OHDA lesions induced PD. High-frequency (130 Hz, 60 µs) SNr-DBS was delivered for 14 days. Locomotor activity (open-field), gait (footprint method), and motor coordination (rotarod) were assessed. Tyrosine hydroxylase (TH) expression in the SN and c-Fos and BDNF expression in the cerebellum, prefrontal cortex (PFC), and ventrolateral thalamus were analyzed histologically. Results: SNr-DBS significantly improved ambulation and horizontal activity compared to the PD group (p < 0.05). Gait analysis showed significant improvements in forelimb/hindlimb stride length and stance width, while rotarod performance indicated enhanced motor coordination (p < 0.05). Histology revealed increased TH expression in the SN and elevated c-Fos and BDNF levels in the cerebellum, PFC, and thalamus in the SNr-DBS group vs. PD rats (p < 0.05). Conclusions: Mid-term SNr-DBS produced significant functional gains in motor activity and coordination in a 6-OHDA PD model, together with molecular evidence of dopaminergic enhancement and neuroplastic activation. These translational findings suggest that targeting the SNr may offer a clinically relevant alternative for patients with PD, particularly for those who may not optimally respond to conventional STN or GPi stimulation. Full article

Background: Osteotomy around the knee (OAK) is a joint-preserving surgery for knee osteoarthritis, yet some patients experience suboptimal outcomes. Preoperative identification of high-risk patients remains challenging. This study aimed to develop a machine learning model to predict clinical outcomes after OAK using preoperative gait acceleration data from inertial measurement units (IMUs). Methods: This multicenter prospective study enrolled patients undergoing OAK. Preoperative gait was recorded using synchronized IMUs placed on the lumbar spine and tibia. Lumbar and tibial signals were used for gait-cycle segmentation, while wavelet-based time–frequency features were extracted from tibial acceleration only. Outcomes were defined by achievement of the minimal clinically important difference in ≥3 KOOS subscales at 2-year follow-up (Good vs. Poor). Continuous wavelet transform features (5–20 Hz) were summarized as mean and standard deviation across six stance subphases. A Random Undersampling Boost classifier was trained and evaluated using nested leave-one-subject-out cross-validation. A sensitivity analysis using logistic regression confirmed that the IMU-based prediction score was independently associated with outcome after adjustment for baseline KOOS (p = 0.047). Results: Of 67 enrolled patients, 37 were classified as Good and 30 as Poor outcome. For machine learning analysis, 1173 tibial acceleration gait-cycle waveforms were usable. The model achieved an AUC of 0.744 (95% CI, 0.610–0.860) using a median of 15 features (range, 5–25) with sensitivity of 0.69 and specificity of 0.72. The most informative predictors were the mean magnitude in the 5–8 Hz band during loading response (0–17%) and variability in the 5–8 Hz band during late stance (67–83%). No significant differences in baseline demographics or radiographic parameters were found between outcome groups. Conclusions: Preoperative IMU-derived gait acceleration features showed moderate-to-good discrimination between outcome groups and may support preoperative risk stratification and individualized perioperative management. Full article

Background: Few studies have examined changes in diet quality into old age, and related these changes to musculoskeletal outcomes. We examined this among Hertfordshire Cohort Study participants. Methods: In total, 178 individuals provided diet quality scores derived in 1998–2004, 2011 and 2017 (median age 64.0, 74.7 and 80.7) using principal component analysis of food frequency questionnaires; higher scores indicated healthier diets (more fruit and vegetables, oily fish and wholemeal bread, and less white bread, added sugar, full-fat dairy products, chips and processed meat). Pearson correlations between diet quality scores at each time-point were computed. Group-based trajectory modelling of diet quality scores was implemented; trajectory groups as predictors of musculoskeletal outcomes (history of hip/knee replacement, osteoporosis, fall in previous year, low grip strength, low gait speed) in 2017 were examined using logistic regression with age and sex included as covariates. Results: Diet quality showed moderate stability over time (0.64 < r < 0.74). Three trajectory groups were identified: low (29%), medium (51%), and high diet quality (20%). A higher diet quality group was related to greater odds (95% CI) of hip/knee replacement (1.85 (1.05, 3.26) per higher category); associations with other musculoskeletal outcomes were weak (p > 0.17). Conclusions: Weak associations were observed between diet quality trajectories and musculoskeletal outcomes. However, higher diet quality was related to increased likelihood of hip/knee joint replacement, potentially due to confounding by socioeconomic position. The stability of diet quality suggests individuals with poorer diets around age 65 are likely to maintain these patterns into old age and may benefit from targeted interventions. Full article

Wearable robots can improve human walking economy; however, their effectiveness depends on user adaptation to assistance. This study introduces a framework for real-time estimation of user adaptation that relies only on wearable sensor data during operation. Metabolic measurements were used solely to establish the ground truth adaptation curves for model training and validation but are not required for real-time inference. Five healthy adults completed six days of treadmill walking while wearing a soft hip exosuit that provided hip extension assistance. Thigh-mounted inertial measurement units recorded step timing and hip-angle trajectories, from which three variability-based features (step-frequency variability, maximum hip-flexion variability, and maximum hip-extension variability) were extracted. A Long Short-Term Memory (LSTM) model used these gait-variability inputs to estimate each user’s adaptation level relative to a metabolic cost benchmark obtained from respiratory gas analysis. Across sessions, the metabolic cost decreased by 9.0 ± 5.6% from Day 1 to Day 6 (p < 0.01) with a mean time constant of 202 ± 78 min, In contrast, the variability in step frequency, maximum hip flexion, and maximum hip extension decreased by 66.4 ± 6.8%, 37.9 ± 24.2%, and 42.8 ± 10.6%, respectively, indicating that these reductions were users’ progressive adaptation to the exosuit’s assistance. Under leave-one-subject-out (LOSO) evaluation across five participants, 59.2% of the model predictions fell within ±10 percentage points of the metabolic cost–based adaptation curve. These results suggest that simple kinematic variability measured with wearable sensors can track user adaptation and support practical approaches to real-time monitoring. Such capability can facilitate adaptive control and training protocols that personalize exosuit assistance. Full article

Population aging is driving the need for unobtrusive, continuous monitoring solutions in residential care environments. Radio-frequency (RF)-based technologies such as Ultra-Wideband (UWB) and millimeter-wave (mmWave) radar are particularly attractive for providing detailed information on presence and movement while preserving privacy. Building on a UWB–mmWave localization system deployed in a senior living residence, this paper focuses on the data-processing methodology for extracting quantitative mobility indicators from long-term indoor monitoring data. The system combines a device-free mmWave radar setup in bedrooms and bathrooms with a tag-based UWB positioning system in common areas. For mmWave data, an adaptive short-term average/long-term average (STA/LTA) detector operating on an aggregated, normalized radar energy signal is used to classify micro- and macromovements into bedroom occupancy and non-sedentary activity episodes. For UWB data, a partially constrained Kalman filter with a nearly constant velocity dynamics model and floor-plan information yields smoothed trajectories, from which daily gait- and mobility-related metrics are derived. The approach is illustrated using one-day samples from three users as a proof of concept. The proposed methodology provides individualized indicators of bedroom occupancy, sedentary behavior, and mobility in shared spaces, supporting the feasibility of combined UWB and mmWave radar sensing for longitudinal routine analysis in real-world elderly care environments. Full article

Background/Objectives: Speech and motor control share overlapping neural mechanisms, yet their quantitative relationships in Parkinson’s disease (PD) remain underexplored. This study investigated bidirectional associations between acoustic voice features and objective motor metrics to better understand how vocal and motor systems relate in PD. Methods: Cross-sectional baseline data from participants in a randomized neuromodulation trial were analyzed (n = 13). Motor performance was captured using an Integrated Motion Analysis Suite (IMAS), which enabled quantitative, objective characterization of motor performance during balance, gait, and upper- and lower-limb tasks. Acoustic analyses included harmonic-to-noise ratio (HNR), smoothed cepstral peak prominence (CPPS), jitter, shimmer, median fundamental frequency (F0), F0 standard deviation (SD F0), and voice intensity. Univariate linear regressions were conducted in both directions (voice ↔ motor), as well as partial correlations controlling for PD motor symptom severity. Results: When modeling voice outcomes, faster motor performance and shorter movement durations were associated with acoustically clearer voice features (e.g., higher elbow flexion-extension peak speed with higher voice HNR, β = 8.5, R² = 0.56, p = 0.01). Similarly, when modeling motor outcomes, clearer voice measures were linked with faster movement speed and shorter movement durations (e.g., higher voice HNR with higher peak movement speed in elbow flexion/extension, β = 0.07, R² = 0.56, p = 0.01). Conclusions: Voice and motor measures in PD showed significant bidirectional associations, suggesting shared sensorimotor control. These exploratory findings, while limited by sample size, support the feasibility of integrated multimodal assessment for future longitudinal studies. Full article

Parkinson’s Disease (PD) is a neurodegenerative disorder in which gait abnormalities serve as key indicators of motor impairment and disease progression. Although wearable sensor-based gait analysis has advanced, existing methods still face challenges in modeling multi-sensor spatial relationships, extracting adaptive multi-scale temporal features, and effectively integrating time–frequency information. To address these issues, this paper proposes a multi-sensor gait neural network that integrates biomechanical priors with time–frequency collaborative learning for the automatic assessment of PD gait severity. The framework consists of three core modules: (1) BGS-GAT (Biomechanics-Guided Graph Attention Network), which constructs a sensor graph based on plantar anatomy and explicitly models inter-regional force dependencies via graph attention; (2) AMS-Inception1D (Adaptive Multi-Scale Inception-1D), which employs dilated convolutions and channel attention to extract multi-scale temporal features adaptively; and (3) TF-Branch (Time–Frequency Branch), which applies Real-valued Fast Fourier Transform (RFFT) and frequency-domain convolution to capture rhythmic and high-frequency components, enabling complementary time–frequency representation. Experiments on the PhysioNet multi-channel foot pressure dataset demonstrate that the proposed model achieves 0.930 in accuracy and 0.925 in F1-score for four-class severity classification, outperforming state-of-the-art deep learning models. Full article

Objective: This study aimed to systematically evaluate the therapeutic effects of repetitive transcranial magnetic stimulation (rTMS) on gait, motor function, and balance in patients with Parkinson’s disease (PD) and identify optimal stimulation parameters for clinical application. Methods: This systematic review and meta-analysis of randomized controlled trials (CTs) was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. PubMed, EMBASE, Cochrane Central, Scopus, and Ovid-LWW were searched until December 2024 for RCTs evaluating the effects of rTMS on PD-related gait, balance, or motor outcomes. Nineteen studies (n = 547) met the inclusion criteria. Data on study characteristics, rTMS protocols (frequency, target area, pulses, session duration, number of sessions, and treatment duration), and outcome measures (freezing of gait questionnaire [FOG-Q], gait speed, Unified Parkinson’s Disease Rating Scale Part III [UPDRS-III], UPDRS total, and timed up and go [TUG] test) were extracted. Effect sizes (Hedges’ g) were pooled using inverse variance meta-analysis, heterogeneity was assessed using I², and publication bias was assessed using funnel plots and Egger’s regression. Results: rTMS produced significant improvements in gait freezing (FOG-Q: g = −0.74; 95% confidence interval [CI] [−1.05, −0.43]; p < 0.001), gait speed (g = 0.62; 95% CI [0.29, 0.95]; p < 0.001), and motor symptoms (UPDRS-III: g = −0.42; 95% CI [−0.70, −0.15]; p = 0.003). No significant effects were observed for UPDRS total (g = 0.18; p = 0.58) or balance (TUG, g = −0.29; p = 0.06). Egger’s test indicated publication bias for gait speed (p = 0.016); however, trim-and-fill imputed zero studies. Subgroup analyses indicated that high-frequency stimulation of the supplementary motor area (SMA) for ≥20 min over 10 sessions (total duration <2 weeks or ≥2 weeks) optimally improved gait speed, whereas low-frequency stimulation targeting M1 and SMA with >1000 pulses per session for 20 min over 10 sessions within <2 weeks most effectively improved the UPDRS-III scores. Conclusions: rTMS exerts moderate and significant benefits on gait and motor performance in PD, particularly when tailored protocols involving SMA or M1 stimulation are employed. High-frequency SMA protocols improve gait speed, whereas low-frequency M1/SMA protocols optimize motor symptom relief. These findings provide evidence-based guidance for rTMS implementation in PD rehabilitation. Full article