Search Results (468)

While classical clustering methods, particularly k-means, produce powerful and practical solutions based on metric distances between data points, they can be limited in complex, nonlinear, and structurally disordered datasets. This study proposes a hybrid topological–metric clustering framework, referred to as Hybrid-NHTSI, that integrates persistent homology-based structural information into the clustering update process. The method is based on the Topological Cluster Separation Index (TCSI), a persistent homology (PH)-based metric for topological separation. In addition to TCSI, the proposed framework uses the Normalized Topological Cluster Separation Index (NTCSI), the Hybrid Topological Clustering Index (HTCI), and the Normalized Hybrid Topological Separation Index (NHTSI) to evaluate clustering performance from both geometric and topological perspectives. In the proposed approach, while the topological separation between clusters is increased, intra-cluster geometric scattering is controlled by a regularization term. This formulation enables the extraction of clusters that are consistent not only topologically but also geometrically. The performance of the method was evaluated on synthetic circles-and-moons benchmark datasets under different noise and overlap levels, and on the UCI Human Activity Recognition real sensor dataset. The experimental results showed that DBSCAN achieved the strongest overall performance on the density-favorable synthetic benchmark, which is consistent with the nonconvex and density-separable structure of the data. However, Hybrid-NHTSI produced higher NTCSI, HTCI, and NHTSI values than classical metric/geometric baselines such as k-means, Spectral Clustering, and Agglomerative Clustering. Pairwise statistical comparisons based on NHTSI confirmed that these improvements were significant against several competing methods. In the real-data experiment, although Spectral Clustering achieved the highest ARI value, Hybrid-NHTSI obtained the highest NTCSI, HTCI, and NHTSI values and significantly outperformed all competing methods in terms of NHTSI. The findings demonstrate that considering both metric and topological information together, rather than relying solely on metric or topological information, provides a more structurally informed evaluation and optimization mechanism for complex clustering problems. Accordingly, the proposed method should not be interpreted as a universally superior clustering algorithm across all metrics, but rather as a topology-aware hybrid refinement framework that enriches metric-based clustering with persistent homology. Full article

The rate control algorithm is designed for natural video by default in video-coding standards. However, computer-generated screen content video (SCV) is very different from natural video captured by a camera, with many different statistical characteristics, such as sharp edges, thin lines, and flat area. This will lead to a difference in the focus of the human visual system (HVS) when viewing on-screen content video. Especially in various sensor data visualization applications such as intelligent display terminals, industrial monitoring and human–computer interaction interfaces, screen content video carries key information collected and reconstructed by image sensors, vision sensors and multimodal sensors. Its edge structures and local details directly affect the interpretation accuracy and application reliability of sensor information. Therefore, it is crucial to investigate perceptual rate control methods that integrate both video content characteristics and human visual perception properties, which possesses substantial theoretical and practical significance. In this paper, we propose a perceptual rate control algorithm for screen content video based on just-noticeable distortion (JND) which is established on the edge profile reconstruction with tolerable variations. First of all, target bit rate allocation for the frame level and CTU level is based on a perceptual weight which is calculated on the JND factor and reconstruction edge character. Secondly, under the constraint of the JND model, an intra rate-distortion (RD) model is established under the constraint of the JND model. The similarity between reference frames and reconstructed frames is taken as feedback in this model. Finally, the proposed rate control algorithm (JND–perceptual rate control (PRC)) is integrated to the existing rate control framework in High-Efficiency Video Coding–Screen Content Coding (HEVC-SCC) for improving the coding efficiency. The experimental results show that the proposed algorithm achieves better bit control precision than the platform, as well as improves the R-D performance of screen content video. In particular, compared with the HEVC-SCC reference software, the coding performance is improved by 3.09 dB on average, the bit rate is saved by 26.51% on average, and the average bit rate mismatch is within 1.159%. Full article

Sensor-based human activity recognition (HAR) plays a fundamental role in healthcare monitoring, sports analytics, and ambient-assisted living. Although deep learning has substantially advanced HAR performance, two practical issues still limit its real-world deployment: (i) the distribution shift caused by changes in users or sensor placements can degrade generalization, and (ii) the quadratic $\mathcal{O}\left(L^2\right)$ complexity of standard self-attention hinders efficient long-sequence modeling on resource-constrained wearable devices. To address these issues, we propose DSHformer, which is an accuracy-oriented HAR framework that combines compact channel–temporal encoding with locality-sensitive hashing (LSH)-based attention. Specifically, DSHformer (i) employs a low-parameter patch-based graph-attention encoder to jointly model latent relationships among sensor channel–temporal dynamics; (ii) introduces a trainable prototype pool together with a multi-layer decomposition network to improve intra-class compactness and inter-class separability via prototype alignment; and (iii) introduces a decomposition-stable LSH-based attention mechanism tailored for HAR, whose core design couples prototype-guided feature decomposition with locality-sensitive hashing to ensure that semantically related tokens remain consistently grouped in the same hash bucket even after decomposition-induced attenuation. The mechanism thereby operates at $\mathcal{O}(LlogL)$ attention complexity on longer sensor sequences. Extensive experiments on five public benchmarks (WISDM, UCI-HAR, PAMAP2, Opportunity, and UniMiB-SHAR) show that DSHformer achieves accuracies of 98.6%, 93.7%, 98.4%, 88.5%, and 96.6%, respectively, achieving competitive or superior performance compared with both Transformer variants and HAR-specific baselines under the adopted benchmark protocols. Ablation studies further confirm the complementary contribution of each component. Full article

The majority of currently available hand kinematic databases have been gathered using expensive marker-based systems or are restricted to a particular gesture-recognition task, failing to capture the dynamic nature of joints when the hand is engaged with an object. To address this gap, we introduce the RGB-based Hand Joint Kinematics (RGB-HJK) dataset, a publicly available collection of continuous, frame-level 3D joint angle trajectories, recorded while ten healthy adults (six male, four female; age $25.8\pm 3.2$ years; BMI $22.8\pm 2.0$ kg/m²) performed five standardized object interaction grasps: Power Grasp (cylindrical bottle), Tripod Grasp (pen), Static Power Hold (smartphone), Precision Pinch (thin paper), and Lateral Pinch (book). Data were collected using a standard RGB camera and the MediaPipe Hands markerless pipeline at $26.95\pm 0.29$ Hz, a rate that was stable across all subjects. Each participant completed five trials for each grasp type. After filtering using active hold, 28,111 validated frames remained, with a 100% detection rate for all 250 trials. Intra-subject repeatability was good (mean SD $\le 7.9°$ across all joint grasp combinations) and inter-subject variability was within the range expected based on normal anatomical diversity. Importantly, kinematic validation of the Index Proximal Interphalangeal (PIP) joint (61.8° ± 18.4°) showed values consistent with ranges reported in previous studies using instrumented gloves and depth sensors. Principal Component Analysis (PCA) confirmed clear linear separability among the five grasp configurations. Unlike existing datasets, the RGB-HJK method does not compromise the natural sense of touch and is free of hardware occlusions, thereby providing an easily accessible ecological baseline. Full article

Efficient irrigation management requires reliable information on soil water content, yet low-cost capacitive sensors often lack proper calibration. This study evaluates the metrological performance of two DF Robot probes, SEN0193 (S1) and SEN0308 (S3), under controlled variations in porous media properties. Glass beads of three size classes (<50 µm, 70–110 µm, and 400–600 µm) were used to simulate fine, medium, and coarse textures. Sensors were tested at four water contents (0, 10, 20, and 30%) and four salinity levels (0, 4, 8, and 16 g NaCl L⁻¹). Results show that the manufacturer-recommended air/water calibration is unsuitable for soils or porous media; calibration should instead be performed under dry and saturated conditions specific to the medium. S1 exhibited stable and homogeneous responses, with intra-unit CV ≤ 2%, but moderate calibration accuracy (R² = 0.68–0.80; RMSE = 8.9–12.9% VWC across textures). S3 showed a wider signal range (80–90% larger than S1), better fit in coarse texture (R² = 0.96; RMSE = 3.5% VWC), but higher unit-to-unit variability (CV = 6–14%) and performance degradation in fine and saline media. Although these sensors cannot provide accurate absolute quantification, their ability to track moisture trends makes them useful for irrigation management, provided calibration accounts for medium texture and salinity. Full article

Flexible strain sensors have attracted considerable attention in gait recognition owing to their ability to adhere directly to the skin near joints and transduce local deformation. In existing work, however, sensor placement and orientation are largely determined by anatomical experience, while multi-channel classification still relies on back-end digital processors, whose power consumption and latency constrain system practicality in wearable scenarios. This paper presents an integrated design path that proceeds from skin-mechanics theory through sensor-layout optimization to analog-domain front-end inference. On the layout side, the lines-of-non-extension (LoNE) theory is employed to convert the selection of sensor attachment angles from empirical judgment into a calculable mechanics problem; guided by the spatial course of LoNE in the ankle and knee regions, the positions and angles of the nine sensors are determined individually—channels perpendicular to the LoNE capture maximum strain, channels offset by 45 degrees supplement non-sagittal-plane information, and a channel aligned along the LoNE provides a near-zero-strain reference. On the circuit side, the mathematical equivalence between the weighted summation of a linear classifier and Kirchhoff’s current law (KCL) nodal current superposition is exploited to map the classification operation onto current aggregation in an analog circuit, yielding an in-sensor computing (ISC) front end in which the nine-channel weighted summation is completed in a single analog step. The sensors are fabricated by screen-printing a liquid-metal–polymer composite conductive ink onto a TPU film substrate, with a gauge factor RSD of $6.8\%$ and a tensile linearity $R^2>0.99$. Using walking, running, and stair descent as verification targets, the analog classifier reaches 99% accuracy at the circuit-level functional-verification stage. On real multi-subject data, it achieves $87.0\%\pm 8.4\%$ accuracy under intra-subject cross-session validation, with an analog-domain inference response faster than $100\mathsf{\mu }\mathrm{s}$. This design path is not bound to a specific joint or sensor material; when the layout methodology is extended to additional joint regions and the circuit architecture incorporates multiple outputs to cover more classification categories, the same workflow remains applicable, offering a promising low-power, lightweight technical solution for wearable motion monitoring. Full article

Industrial sensor systems are increasingly vulnerable to both physical anomalies and cyberattacks, while their collected time series typically present complex periodic and non-stationary characteristics, along with dynamic spatial dependencies among sensors. To address these issues, this paper proposes a dual-branch graph learning framework with frequency gating for simultaneous industrial sensor anomaly and cyberattack detection. The model first divides the input time series into multiple patches and decomposes each patch into periodic and non-stationary components via frequency analysis. Two graph isomorphism network branches, namely periodic GIN (P-GIN) and non-stationary GIN (NS-GIN), are designed to model the spatial dependencies of the two components separately, where the graph structure is adaptively learned using a Gaussian kernel-based mechanism. Furthermore, a frequency gating module is introduced in the non-stationary branch to enhance the representation of abnormal and attack-related features. Hierarchical temporal encoding is performed via intra-patch attention and inter-patch attention to capture both local and long-range temporal dependencies. Extensive experimental results on real-world industrial sensor datasets demonstrate that the proposed method achieves superior performance compared with state-of-the-art methods in both anomaly detection and cyberattack detection tasks. Full article

Background: Developmental dysplasia of the hip (DDH) is the most common congenital musculoskeletal disorder in newborns. Flexion–abduction orthoses are widely used in early treatment; however, in vivo data on their biomechanical load characteristics remain limited. This study aimed to evaluate axial force transmission in a hip flexion–abduction orthosis and to compare load patterns between healthy newborns and infants with DDH. Methods: In this exploratory observational study, 36 newborns (19 healthy, 17 with unilateral DDH) were examined within the first week of life. Axial forces transmitted through a Mittelmeier–Graf hip flexion–abduction orthosis (MGO) were measured using integrated force sensors under symmetrical and asymmetrical adjustment configurations. Intergroup comparisons were performed using non-parametric statistical tests. Results: Mean axial forces were significantly higher in healthy infants than in those with DDH under both symmetrical (4.02 N vs. 2.51 N; p = 0.019) and asymmetrical (3.67 N vs. 1.83 N; p = 0.001) conditions. Relative load corresponded to approximately 11–12% of body weight in healthy infants and 5–7% in the DDH group. No significant intra-individual differences were observed between dysplastic and contralateral hips. Orthosis configuration (symmetrical vs. asymmetrical) did not significantly affect load distribution. Conclusions: This exploratory in vivo study demonstrates that axial load transmission in a hip flexion–abduction orthosis is low and influenced by underlying hip pathology. Infants with DDH generate lower forces than healthy newborns, potentially reflecting altered biomechanics. As no significant differences were observed between orthosis configurations, symmetrical adjustment may be favored in clinical practice due to better usability and compliance. Further studies with larger cohorts are needed to confirm these findings. Full article

Background: This study evaluated the reliability and reproducibility of biomechanical variables from a foot-mounted inertial measurement unit (IMU; Stryd^®) and a trunk-mounted triaxial accelerometer (TTA; RunEASI^®) during treadmill running. Methods: Thirteen trained male runners (age: 27.6 ± 8.2 years; body mass: 64.4 ± 6.1 kg; height: 174.2 ± 6.4 cm) completed four treadmill sessions separated by seven days, each with four 5-min trials at 13 km·h⁻¹. Reliability and reproducibility were assessed using intraclass correlation coefficients (ICCs), standard error of measurement (SEM), smallest worthwhile change (SWC) and coefficient of variation (CV). Results: Both devices showed excellent intra-session reliability (ICC ≥ 0.90). Inter-session reproducibility was high for the IMU in running power, step frequency, and ground contact time (ICC = 0.906–0.992; CV = 0.31–1.50%); the TTA showed excellent reproducibility for the spatiotemporal variables—step frequency, ground contact time, and flight ratio (ICC = 0.931–0.956), whereas impact-related variables showed only good reproducibility (ICC = 0.823–0.871; CV up to 7.76%). Mean step frequency values were similar between devices, but trial-to-trial agreement was limited (r = 0.447; R² = 0.20; Bland–Altman bias = 0.70 steps·min⁻¹, LoA: −15.08 to 16.47), and ground contact time showed systematic discrepancies, with the TTA underestimating values (bias = 16.86 ms, LoA: −45.64 to 11.92). Conclusions: Both sensors provide consistent measurements within their own systems; however, differences in sensing location produce different absolute values, so they should not be used interchangeably and interpretation must remain device-specific. Full article

Classifying encrypted sensor traffic is critical for the security and management of Internet of Things networks, particularly in Mobile Edge Computing (MEC) environments. Existing methods often require extensive task-specific labeled data to adapt to emerging traffic categories and may also fail to distinguish intrinsic traffic behaviors from patterns introduced by shared communication libraries, which can degrade classification accuracy under distribution shifts. To address these issues, we propose CDTF, a contrastive dual-task framework for transferable and few-shot traffic representation learning. CDTF adopts a hybrid pre-training strategy that jointly optimizes supervised triplet pretraining (STP) and self-supervised dynamic burst masking (DBM). STP uses base-class labels as structural anchors to explicitly constrain distance relationships by aligning intra-class samples and separating inter-class samples, thereby mitigating interference from shared network components. DBM models global semantic structures and enhances the robustness of traffic representations against network noise and distribution shifts. By learning discriminative and contextual representations in a shared embedding space via these two tasks, CDTF can rapidly adapt to novel categories through lightweight fine-tuning, thereby substantially reducing the reliance on large-scale fine-grained supervision in downstream tasks. Experimental results across seven public and two custom datasets, across diverse environments, show that the proposed framework outperforms state-of-the-art methods. Under the few-shot setting, CDTF improves Precision by 4.61 percentage points over the strongest baseline, with statistical significance confirmed by a paired t-test ($p<0.05$). Full article

Wireless sensor networks (WSNs) are fundamental to the Internet of Things (IoT) and are widely used in environmental, industrial, and healthcare applications. However, their operational lifetime is constrained by the limited energy resources of sensor nodes. The Low-Energy Adaptive Clustering Hierarchy (LEACH) protocol reduces energy consumption through clustering but suffers from random cluster head (CH) selection, leading to uneven energy usage and reduced stability. This study introduces a hybrid optimization approach, LEACH-CSA, which integrates the Crow Search Algorithm (CSA) with LEACH to enhance CH selection and positioning. The proposed method employs CSA’s intelligent search behavior to minimize intra-cluster distances and balance energy consumption across nodes. MATLAB simulations with 100 sensor nodes in a 100 × 100 m² area demonstrate that LEACH-CSA significantly reduces energy consumption and extends network lifetime compared with LEACH and its variants. Furthermore, CSA parameters were optimized using a progressive randomized tuning strategy with 1000, 2000, and 4000 candidate configurations. A comparative evaluation against LEACH-based GA, PSO, GWO, and WOA demonstrated that LEACH-CSA consistently improved the FND metric under different node density and area-scaling scenarios. Full article

Osteoarthritis (OA) and osteochondral degeneration present a significant clinical burden characterized by the complex interplay of extracellular matrix degradation and chronic inflammation. While biochemical profiling has matured, a critical translational gap remains in transitioning from benchtop assays to systems capable of continuous, intra-articular monitoring. This review provides a comprehensive synthesis of experimentally validated biosensing technologies, including optical, electrochemical, and piezoelectric Quartz Crystal Microbalance (QCM) platforms, evaluated through the lens of sensing architecture, biomarker specificity, and matrix compatibility. Our analysis reveals that while optical sensors offer superior sensitivity, electrochemical platforms show the greatest promise for miniaturized, implantable integration. However, a pivot in the field is identified: the primary bottleneck has shifted from analytical detection limits to operational stability within the hostile synovial environment. Current research is largely restricted to single-analyte detection in simplified media, failing to address the multifactorial nature of OA. We propose that the next generation of osteochondral diagnostics must prioritize multiplexed arrays, mechanically compliant architectures, and machine-learning-assisted signal processing. By bridging these engineering frontiers, biosensors will evolve from passive diagnostic tools into intelligent, personalized platforms for real-time disease management. Full article

Background: Accurate three-dimensional positioning of dental implants is critical for ensuring biomechanical stability, prosthetic passivity, and long-term clinical success. While computer-assisted navigation systems achieve high precision, their complexity and cost often limit accessibility. This study presents the development and preclinical experimental validation of a low-cost prototype designed to enhance angular accuracy in dental implant placement within a controlled 3D haptic simulation environment. Methods: A preclinical experimental design was implemented using a 3D haptic simulator (Virteasy, Montpellier, France). The prototype incorporated high-precision inertial measurement units (IMUs) and an Extended Kalman Filter (EKF) for real-time angular feedback. Ninety-seven simulated implant placements were performed—both freehand and with prototype assistance—under identical virtual conditions by a single experienced operator. Angular deviations in mesiodistal and buccolingual planes were recorded, combined into a composite 3D index, and analyzed using paired t-tests and linear mixed-effects models. The study was conducted in a controlled simulation environment, which does not fully replicate clinical conditions. Results: The prototype significantly reduced angular deviation from 13.49° to 2.99° in the mesiodistal plane (−77.8%) and from 13.56° to 5.59° in the buccolingual plane (−58.8%), achieving an overall 67% improvement in three-dimensional orientation (p < 0.001; Cohen’s d = 1.47). Agreement with an optical reference system (OptiTrack) was excellent (bias = +0.36°, RMSE = 0.39°). Intra-operator reliability exceeded 0.95 (ICC), confirming strong reproducibility and measurement stability. Conclusions: The proposed inertial sensor-based prototype achieved angular accuracy within the range reported for computer-guided systems while maintaining advantages of portability, low cost, and usability. Its integration into haptic simulators provides a valid tool for both educational and preclinical applications, offering real-time feedback that enhances spatial perception and psychomotor learning. Future clinical studies should validate its performance in cadaveric and patient-based contexts to determine its practical impact on surgical precision and implant success. Full article

Electrical monitoring of brain activity can be performed discreetly and continuously over long periods of time using intra-auricular electroencephalography (intra-auricular EEG), a promising technique suitable for subjects who are difficult to monitor, such as newborns or patients with neurological conditions requiring discreet but long-term neurophysiological assessment. The concept of intra-aural EEG can be realized through the development of systems that include wearable sensors, whose performance critically depends on the development of biocompatible electrode materials that exhibit low impedance and can maintain and provide stable contact between the electrode and the epithelial tissue. Based on our previous work on carbon nanotube (CNT)-based hydrogel composites for intra-aural EEG electrodes, this study focuses on the electrochemical characterization of hydrogels initially prepared from gelatin methacrylate (GelMA)/2-hydroxyethyl methacrylate (HEMA) doped with varying concentrations of CNTs (0–3 wt%). In the present study, the materials obtained in the first stage were evaluated using electrochemical impedance spectroscopy (EIS) under both liquid and dry conditions, supplemented by measurements of hydration capacity. The results show that the composite with 3% CNT content exhibits suitable properties, making the material making the 3 wt% CNT formulation a promising platform for the further development of 3D-printable hydrogel electrodes for intra-aural EEG applications. Equivalent circuit modeling reveals improved ionic and electronic conductivity compared to the undoped hydrogel, attributed to better CNT dispersion and polymer crosslinking. This work provides insights into the structure–property relationships of CNT–hydrogel composites and lays the foundation for the further development of a 3D-printed and in vitro/in vivo validated prototype of intra-aural EEG sensors. Full article

Accurate Land Use and Land Cover (LULC) mapping in high-latitude mountain regions faces critical challenges from persistent cloud cover and complex topography, which limit the utility of passive optical sensors. To address the absence of evidence-based guidelines for these data-scarce environments, this study employs a systematic ablation design to quantify the marginal and synergistic contributions of optical data (Sentinel-2), Synthetic Aperture Radar (Sentinel-1 SAR), topography, and intra-seasonal phenological metrics within the Aysén River basin, Chilean Patagonia, developing a geospatial workflow with high transferability potential. Using a Random Forest classifier, five progressive configurations were compared: a seasonal optical baseline (A), and configurations incorporating intra-seasonal percentiles (A + P), topography (A + T), SAR (A + R), and their full integration (A + P + T + R). The baseline model achieved an Overall Accuracy (OA) of 89.2% and a Macro-F1 of 80.5%; the fully integrated model reached OA = 92.5% and Macro-F1 = 86.0%. Macro-F1 was adopted as the primary metric because it assigns equal weight to all 11 classes regardless of spatial prevalence, capturing gains in minority but ecologically critical classes that OA would mask. SAR and topographic variables were the largest contributors, generating non-redundant improvements in structurally complex and relief-conditioned classes, respectively. Furthermore, annual SAR composites demonstrated superior cartographic spatial consistency over seasonal aggregations, which introduced purely cartographic geometric artifacts at class ecotones despite achieving marginally higher point-based statistical metrics, a divergence explained by the spatial blindness of confusion-matrix validation to boundary-zone classification errors. Full article