Search Results (631)

This study proposes a multi-source perceptual information fusion method to improve the accuracy and stability of dairy cow behavior monitoring. Existing machine vision approaches are often affected by lighting conditions, occlusion, and complex cowshed environments, while single wearable inertial measurement unit (IMU) devices may confuse similar behaviors such as eating, ruminating, standing, and lying. To address these limitations, a wireless collar was developed to synchronously collect nine-axis IMU data and ultra-wideband (UWB) ranging data in real time. Combined with manual behavioral observations, a dataset covering seven behaviors—eating, ruminating, standing, lying, drinking, sleeping, and lateral trunk contact—was constructed. By integrating neck-motion features extracted from the IMU data with spatial-distance features obtained from the UWB data, an IMU–UWB dual-branch fusion model was developed to automatically classify dairy cow behaviors. The results indicate that the proposed method can effectively reduce confusion among similar behaviors and improve the recognition of behaviors with limited samples. This approach enables more comprehensive assessment of dairy cows’ daily activities and health status, providing technical support for health monitoring, early disease warning, and intelligent dairy farm management. Full article

Upper-limb motor dysfunction resulting from neurological disorders severely limits patients’ activities of daily living and social participation. Pneumatic upper-limb rehabilitation robots have emerged as a promising intervention owing to their inherent compliance, lightweight design, and high power-to-weight ratio, which facilitate safe, repetitive, and home-based training. Despite these advantages, extensive clinical translation remains hindered by challenges including actuator hysteresis, nonlinear dynamics, limited accuracy in intention recognition, and inconsistent clinical evaluation metrics. This review systematically examines recent advancements in pneumatic upper-limb rehabilitation robots across four critical dimensions: structural design, human–robot interaction, control strategies, and clinical translation. We comparatively analyze rigid exoskeletons, soft wearable devices, and rigid–soft hybrid configurations based on output capability, motion accuracy, comfort, and clinical applicability. The findings suggest that while rigid systems offer high precision and soft systems maximize safety, rigid–soft hybrid architectures represent a critical developmental trend for balancing motion accuracy with interaction compliance. Furthermore, the review evaluates multimodal sensing techniques (e.g., EMG, EEG, and IMUs) for motion intention decoding and training state monitoring, alongside conventional, adaptive, and artificial intelligence-driven control methods aimed at compensating for pneumatic nonlinearity and improving real-time response. Current clinical evidence indicates that these systems effectively enhance upper-limb function and muscle strength, particularly in post-stroke rehabilitation; however, existing trials are frequently constrained by small sample sizes, short interventions, and heterogeneous protocols. Future research must prioritize rigid–soft hybrid architectures, robust multimodal sensor fusion, digital twin-assisted assessment, adaptive intelligent control, and standardized home-based rehabilitation platforms. Ultimately, this comprehensive review provides a concise reference for the design optimization and clinical deployment of next-generation pneumatic rehabilitation systems. Full article

Imaging underpins the diagnosis and structural staging of osteonecrosis of the femoral head (ONFH) but does not directly quantify functional impairment during weight-bearing walking. We developed a subject-level ONFH classification framework using multilevel gait features acquired with wearable inertial measurement units (IMUs). Thirty healthy controls and 21 participants with imaging-confirmed ONFH completed self-paced walking trials recorded at 100 Hz. Gait cycles were segmented from bilateral foot-contact events, normalized to 120 points, and represented as 17-channel kinematic waveforms, 22-dimensional cycle-level scalar features, and 7-channel dynamic absolute asymmetry waveforms. These inputs were encoded by CNN–CBAM–BiLSTM, multilayer perceptron, and one-dimensional convolutional branches, respectively, and fused at the feature level. Evaluation used 51-fold leave-one-subject-out cross-validation, training-fold-only preprocessing, within-subject probability averaging, and five predefined random seeds. The five-seed ensemble achieved an accuracy of 0.9412, sensitivity of 0.8571, specificity of 1.0000, F1-score of 0.9231, and area under the receiver operating characteristic curve of 0.9556. Ablation analysis identified the scalar-feature vector as the principal source of incremental performance; the dynamic asymmetry branch contributed complementary information only in the complete model. These findings provide preliminary evidence for further evaluation of wearable gait-based ONFH classification in independent cohorts and objective functional assessment. Full article

Wearable inertial measurement units (IMUs) offer an accessible alternative to optical motion capture (MoCap) gait analysis, but their performance in Progressive Supranuclear Palsy (PSP) requires validation. We assessed the concurrent validity of IMU-derived versus MoCap-derived gait metrics and static postural sway in 30 patients with PSP using Bland–Altman analysis, Intraclass Correlation Coefficients (ICC), and Spearman rank correlations. Finally, we assessed equivalence using the Two one-sided tests (TOST) procedure. Multivariable linear regression was used to determine whether clinical severity, as measured by the PSP Rating Scale (PSPRS), independently predicted absolute IMU measurement error while controlling for patient age and gait velocity. IMUs demonstrated excellent between-system agreement for parameters such as cadence (100.76 ± 11.42 vs. 100.52 ± 11.59) and cycle time (1.21 ± 0.15 vs. 1.22 ± 0.15; ICC > 0.98), despite a systematic underestimation of gait velocity (p < 0.05). Agreement significantly diminished for micro-phases (e.g., single/double support times) and spatial asymmetry. Interestingly, the TOST procedure revealed that only sagittal and transverse trunk kinematics were equivalent between systems, with all other measures failing to find equivalency. For static sway, the IMU demonstrated strong rank-order correspondence for tracking relative postural instability (ρ = 0.82, p < 0.05). Multivariable analysis revealed that higher PSPRS scores are independently associated with greater between-system discrepancies in support phases and pelvic and trunk kinematics (p < 0.05), irrespective of reduced gait speed. These findings highlight the need to develop disease-specific algorithms, rather than relying on normative commercial models, to establish reliable digital biomarkers for monitoring progressive motor decline. Full article

Accurate assessment of joint kinematics is fundamental to biomechanics, rehabilitation, and sports science. Although optical motion capture (OMC) remains the laboratory reference standard for biomechanical validation, its cost, infrastructure requirements, and limited applicability outside controlled environments restrict its broader use. Wearable inertial measurement units (IMUs) and vision-based markerless systems have consequently emerged as complementary alternatives, offering portability, reduced subject preparation, and applicability in ecological settings. Their rapid development, however, has not always been accompanied by an equally rigorous metrological interpretation of performance. This narrative review provides a comparative analysis of IMU- and vision-based approaches for joint kinematics estimation, focusing on biomechanical validation metrics and measurement error. Because the primary literature reports fundamentally different quantities under widely differing experimental conditions, evidence is presented stratified by outcome class, joint, plane of motion, task, and acquisition dimensionality, and values belonging to different outcome classes are not pooled. For sagittal-plane lower-limb angles during level walking in healthy adults, with careful sensor-to-segment calibration and an optoelectronic reference, IMU-based systems show the most consistent performance, with RMSE commonly between 3° and 6°. Vision-based systems achieve comparable accuracy for selected outcomes, particularly spatiotemporal gait parameters and sagittal-plane angles in controlled views, while degrading with occlusion, motion blur, and depth ambiguity. Accuracy is therefore not an intrinsic property of the sensing modality but of the entire measurement chain, including calibration, biomechanical modeling, acquisition geometry, and reporting conventions. Rather than ranking technologies by accuracy alone, the measurement requirements should be derived from the intended application. Full article

Robotic teleoperation is a core enabling technology spanning remote surgery, industrial inspection, and virtual-reality applications. Despite growing deployment, operator experience assessment still relies almost exclusively on post hoc subjective questionnaires, which preclude real-time monitoring and adaptive intervention. Here, the wearable IMU chain is considered not only as a command interface but also as an implicit sensing channel for operator state. We test whether end-effector kinematics generated by the IMU-to-robot mapping contain information about ten post-task workload and user-experience dimensions, comprising NASA-TLX-inspired workload scales together with usability, responsiveness, realism, intuitiveness, and perceived performance. A secondary analysis of a publicly available dataset (16 participants, 144 motion recordings, simulated UR10e arm) is conducted through a three-stage pipeline: bivariate correlation analysis (Pearson and Spearman), multivariate regression (10 model families, 16 feature-set combinations, Leave-One-Subject-Out validation), and binary classification (median-split). Statistical validity is assessed via 1000-permutation nested testing. Target-specific regression models reach $R^2\ge 0.50$ on seven out of 10 subjective dimensions, with a peak of $R^2=0.787$ for usability; permutation testing confirms significance for eight out of 10 targets. Binary classification achieves AUC $\ge 0.75$ on nine out of 10 targets, with three dimensions reaching perfect AUC. SHAP analysis identifies temporal irregularity and distributional shape descriptors as the dominant kinematic explanatory families. These results support the feasibility of kinematics-based inference of operator experience and provide an offline proof of concept toward future real-time adaptive teleoperation systems. Full article

Background: Quantitative assessment of dynamic balance remains challenging in clinical practice. Wearable inertial measurement unit (IMU)-based technologies offer an objective and accessible approach for monitoring postural control. This study aimed to develop and preliminarily validate a wearable IMU-based trunk sway monitoring system (SwayTracker) and investigate trunk sway characteristics in stroke patients walking with and without arm sling support. Methods: A sternum-mounted IMU system was developed to quantify dynamic trunk sway during walking. Fifteen healthy adults and fourteen stroke patients participated. The healthy participants established normative reference values, whereas the stroke patients completed walking trials with and without arm sling support. Trunk sway was quantified using anteroposterior (AP) and mediolateral (ML) deviations and a polar-coordinate-based sway model. Results: The healthy reference cohort exhibited a mean trunk sway magnitude (radius) of 8.20 ± 3.16°. The stroke patients demonstrated greater sway during unsupported (15.36 ± 5.83°) and sling-assisted walking (15.35 ± 4.59°). Although overall sway magnitude remained unchanged, the mean sway direction shifted from 69.49° to 110.20°, indicating a redistribution of trunk sway from the anterior-right toward the anterior-left quadrant. Forward sway remained the dominant AP component, whereas ML sway shifted from predominantly rightward to leftward with sling use. Conclusions: SwayTracker provides a feasible method for objective assessment of dynamic trunk sway during walking. The stroke patients exhibited increased sway magnitude and altered directional organization compared with healthy individuals. Arm sling use primarily modified ML postural compensation patterns rather than reducing overall trunk sway, highlighting the potential of wearable trunk sway monitoring for gait and balance assessment in neurological rehabilitation. Full article

Stroke and other neurological injuries impair hand function. Although rehabilitation therapists encourage reintegration of the affected hand into daily activities, there are few tools that can be worn during everyday life that provide quantitative feedback on how much or how effectively the hand is used. Wearable sensors that can accurately track hand movements and are easily applied and removed can present intuitive feedback that could motivate hand use similarly to how pedometers encourage walking. While tracking all the hand and finger joints is needed for scientific studies, under-sensorization, or using fewer sensors than required for tracking all degrees of freedom, may suffice for providing users feedback about hand use in everyday life. Further, it may enable a wearable device to be more easily donned and doffed, more power efficient, and more cost efficient. Here we develop a low-cost, multi-sensor, wireless wearable system for tracking selected hand and wrist movements during everyday life. The system includes fabricated soft, stretchable “e-skin” bend sensors and off-the-shelf inertial measurement units (IMUs) that accurately measure finger bend angles and wrist movements. The system also includes an application and removal protocol that enabled naïve unimpaired participants to apply and remove the system in ~5 min and ~4 min, respectively. The system cost was $111 per device, with prices falling to an estimate of $55 when manufactured at scale. This hand wearable demonstrates accurate kinematic tracking and user-friendly donning/doffing workflows for unimpaired participants, making it a promising platform for everyday hand tracking. Future work will extend this platform to the movement-impaired population for neurorehabilitation applications. Full article

Gait analysis is a well-known tool used to evaluate human locomotion. This analysis is usually performed in movement analysis laboratories equipped with optoelectronic cameras. However, many wearable devices based on the use of Inertial Measurement Units (IMUs) have been developed to overcome the limitations of the gold standard technology. LUBU Technologies developed insoles equipped with IMUs and algorithms for the extraction of gait parameters such as gait events, temporal parameters and stride length. The aim of the present study was the validation of these parameters against an optoelectronic reference system. Twenty people were involved in performing a series of indoor walking trials to assess the accuracy of the insole measurements and an additional group of ten people were asked to walk outdoors to preliminarily assess total distance estimation under ecological walking conditions. The results showed low errors and good agreement for all the parameters computed by the insoles. Moreover, the outdoor assessment offered preliminary insights into performance in a more ecological setting. Overall, the findings suggest that LUBU insoles may represent a valid wearable tool for estimating gait spatio-temporal parameters in healthy young adults. However, further investigations are required in older adults or populations with pathological gait patterns. Full article

Wearable inertial sensing and markerless vision are increasingly integrated to enable objective assessment of locomotor and postural function for sports telerehabilitation, intelligent physiotherapy, and athlete performance monitoring. Before such multimodal systems can be translated to clinical practice, their fusion, optimization, and temporal modeling strategies require validation under controlled conditions with reliable ground truth. This study presents a unified multimodal framework that hierarchically integrates inertial measurement unit (IMU) signals and RGB visual information through kernelized representation learning, adaptive multimodal fusion, evolutionary feature optimization, and deep temporal classification. The IMU branch employs Kernelized Extreme Learning Machine (KELM) denoising, Kernelized Canonical Correlation Fusion (KCCF), entropy-guided adaptive windowing, and complementary time-series descriptors (MINIROCKET, TS-CHIEF, and r-STSF). Concurrently, the RGB branch combines anisotropic diffusion filtering, HRNet-based silhouette extraction, DensePose R-CNN, Mesh Graphormer, and Multi-Model Pose-Flow Fusion (MPFF) to learn robust visual representations. Both modalities are integrated through Weighted Canonical Feature Fusion (WCFF) and optimized using a Genetic Algorithm for feature selection and adaptive modality weighting before temporal modeling with cluster-based alignment, Gaussian Process Sequence Modeling, and DeepConvLSTM. As the selected benchmarks do not provide complete inertial recordings, the inertial modality is established according to the adopted experimental protocol to support multimodal fusion analysis. Under 5-fold subject-independent cross-validation, the framework achieves accuracies of 86.56 ± 0.31% on SoccerDiffusion and 88.04 ± 0.25% on HumanoidRobotPose. Although evaluated on humanoid robotic benchmarks, the proposed framework provides a methodological basis for future wearable-enabled clinical movement assessment, remote rehabilitation, and athlete monitoring, while validation on synchronized human inertial-visual datasets remains an important direction for future research. Full article

In Internet-of-Things (IoT)-based rehabilitation, patient data from wearable sensors (IMU, EMG, heart rate), clinical assessments, and surveys differ in format and granularity, complicating unified patient-state construction. Existing AI-based systems often omit fatigue level or rehabilitation stage, or they place large language models (LLMs) in the clinical decision-making role, exposing patients to hallucination risk. We propose an IoT-ready framework comprising three components: an input-source-independent context abstraction pipeline, a digital twin that simulates clinical-score trajectories, and an LLM Agent that interprets the outputs of a deterministic algorithm and a digital twin as (subject, predicate, object) Triplets to produce a retrieval-augmented clinical report. The algorithm—not the LLM—selects the 13-exercise sequence from Shoulder Pain and Disability Index (SPADI) item-level responses. Explicit per-layer schemas let IoT-sensor branches be added without changing the downstream interface. Here we evaluate the framework on the clinical-score path (three patient-reported outcome measures and six range-of-motion measures); the multi-modal IoT branches maintain deployment-target functionality. On 48 IRB-approved shoulder rehabilitation patients (144 longitudinal records; augmented to 7200 only to train the trajectory generator), real-only leave-one-subject-out evaluation gave VAS RMSE 0.658 (0–10) and SPADI RMSE 6.499 (0–100). Ablation across four surface forms revealed a fidelity–accuracy trade-off—Narrative highest on template fidelity (BERTScore F1 0.250), raw JSON highest on judge-rated accuracy—and the framework adopts the balanced-midpoint Triplet form. A claim-level audit of the Triplet-form reports left 22–31% of atomic claims unsupported (Claude Sonnet and GPT-4o judges) regardless of guideline retrieval. A raw-LLM control never reproduced the algorithm-defined sequence exactly (0 of 48), supporting deterministic, auditable sequence selection and the need for clinician review before clinical use. Full article

Wearable inertial measurement units (IMUs) offer an objective, low-cost basis for Parkinson’s disease (PD) assessment, but multi-task clinical protocols yield heterogeneous recordings across body locations and small cohorts, and it is unclear whether such data can support reliable PD detection without training deep models from scratch. We therefore ask whether a motion-pretrained representation transfers to this setting, and quantify how much of the discriminative signal it supplies. Each subject is represented by five task-level motion embeddings, one per clinical task, produced by a frozen pretrained spatio-temporal graph convolutional network (ST-GCN) that fuses the thirteen body-worn sensors into a whole-body embedding; a three-layer Transformer with validity-mask weighting aggregates these tokens for binary PD-versus-control classification on the WearGait-PD cohort (181 subjects: 100 PD, 81 controls). Under a leakage-free nested protocol with repeated subject-disjoint stratified 5-fold cross-validation (5 seeds; 25 estimates per model) and paired significance testing, the model attains a balanced accuracy of 0.834 ± 0.087, macro-F1 of 0.842 ± 0.094, and AUC of 0.842 ± 0.103. It leads six classical baselines and a spectrogram-CNN on accuracy-based metrics, though random forest, gradient boosting, and the spectrogram-CNN edge ahead on AUC; after correction for fold correlation, none of these between-model differences is significant. The one robust finding is a transfer effect: replacing the pretrained encoder with a random one of identical architecture lowers balanced accuracy by 15.5 points when frozen (p = 0.043) and 20.4 when trained end-to-end (p = 0.014). Discrimination is preserved under 1:1 age matching (0.846) and across both genders, so it is not explained by age imbalance. Motion-pretrained skeletal encoders thus supply the majority of the discriminative signal, while the aggregator contributes gains inseparable from noise at this cohort size. Full article

Precision livestock farming (PLF) is hindered by high costs, infrastructure demands, and complex deployment. To address these barriers, we developed CABRA, an open-hardware wearable system for real-time behaviour monitoring in pasture-based livestock. The collar-mounted device integrates a 6-axis IMU, a GPS, and a low-power ESP32 microcontroller within a modular architecture, using the routerless ESP-NOW protocol to transmit data directly to a base station—eliminating reliance on network infrastructure or cloud connectivity. The system supports both synchronised data logging for video annotation and real-time embedded behaviour classification via an optimized decision-tree pipeline deployed directly on the microcontroller. Field trials with dairy goats confirmed robust hardware performance, minimal animal disturbance, and reliable communication over 100 m. A two-stage evaluation revealed that while the extracted IMU features are highly discriminative (achieving F1 > 0.99 under window-level validation), cross-animal generalization remains challenging (macro F1 = 0.31 under rigorous animal-level partitioning), primarily due to the “sensor placement effect” and domain shift between individuals. These results honestly quantify the current limitations of uncalibrated wearable livestock sensing while validating the functional feasibility of edge-based inference. All design assets—CAD files, schematics, firmware, and data pipelines—are openly released to ensure full reproducibility and community-driven adaptation for diverse PLF applications. Full article

Gait symmetry analysis has emerged as an important tool in rehabilitation engineering and neurological disorder assessment, as it provides clinically relevant indicators of mobility impairment and gait abnormalities. The proposed framework integrates gait symmetry variability, statistical gait features and Dynamic Time Warping (DTW)-based temporal alignment to enhance robustness against gait variations and irregular walking patterns. A hybrid feature vector comprising DTW similarity scores, the hybrid gait symmetry index (GSI), and statistical gait descriptors was employed to classify gait patterns into five categories: normal, slow, medium, fast, and abnormal. The system was implemented as a wearable edge-computing platform using an NI myRIO device equipped with a tri-axial Inertial Measurement Unit (IMU) mounted on the subject’s body. The onboard FPGA performs real-time signal preprocessing, GSI computation, feature extraction, constrained DTW matching, and gait classification using fixed-point streaming architectures and BRAM-based buffering. Meanwhile, the embedded ARM processor manages TCP/IP communication and transmits real-time gait information to a remote monitoring workstation via a WiFi interface for visualization and analysis. Operating at a clock frequency of 100 MHz, the complete architecture achieves an end-to-end processing latency of approximately 4 ms. The proposed FPGA-based implementation provides low-latency, energy-efficient, and real-time gait analysis, making it well suited for wearable rehabilitation systems, assistive healthcare devices, and continuous mobility monitoring applications. Full article

Portable biosensor hardware can now sustain continuous multimodal physiological acquisition at the edge, yet the analytical layer that converts raw signals into deployment-consistent inference remains the main bottleneck for practical embedded systems. This study addresses that bottleneck by presenting the machine-learning layer of the Real-time Cognitive Grid, the analytical companion to the previously reported hardware architecture, which equips a fixed-wiring biosensor assembly with real-time physiological-state classification through an asymmetric edge–cloud workflow. The proposed framework assigns analytical responsibility across tiers: a locked 17-feature schema comprising 5 EMG features, 6 EEG spectral features, 2 cross-modal features, 2 HRV features, 1 EOG feature, and 1 EEG quality indicator governs window-bounded inference on the Arduino Nano RP2040 Connect with an LDA edge artefact requiring approximately 716 B RAM, whereas the cloud tier supports public-dataset pretraining, hardware-aligned refinement, multimodal fusion, deployment comparison, and feature-importance analysis under the same schema contract. To evaluate analytical consistency across physiological diversity, five public repositories covering stress physiology (WESAD), affective EEG (DEAP), inertial activity recognition (PAMAP2), sEMG gesture decoding (EMG Gestures), and motor-imagery EEG (EEGMMIDB) were evaluated under subject-disjoint GroupKFold (k = 5) protocols. To test whether the same contract survives translation to the physical rig, the hardware branch was evaluated under session-disjoint GroupKFold across five bench-acquired sessions. Unimodal performance was strongest in sEMG- and IMU-dominant tasks, whereas multimodal fusion improved macro-F1 by up to 0.141 over the strongest unimodal baseline in WESAD and by 0.109 in PAMAP2. In the hardware branch, the deployed edge LDA artefact reached 0.9435 macro-F1 with 0.9470 accuracy, while the retained cloud Random Forest reached 0.8792 macro-F1 with 0.8799 accuracy; feature-importance analysis further showed that the final 17-feature branch was dominated by EMG descriptors, with EEG spectral terms contributing secondary support and hardware-exclusive variables remaining weak under the present bench regime. These results show that a compact multimodal sensing assembly can be elevated beyond passive signal capture into an intelligent portable biosensor that performs context-aware interpretation with minimal user intervention, supported by a reproducible analytical workflow that remains coherent across heterogeneous benchmark repositories, hardware-specific refinement, and microcontroller-class deployment, thereby establishing cross-session bench feasibility as a structured basis for future multi-subject wearable validation. Full article