Search Results (151)

Background/Objectives: Knee osteoarthritis (KOA) is associated with pain, functional decline, and altered biomechanics. The Step-Up and Down-Forwards (StUD-F) task provides an ecologically relevant assessment of challenging movements. This study investigated neuromuscular activation and lower-limb kinematics of leading and trailing-limbs during the StUD-F in individuals with KOA. Methods: Forty participants with KOA (65.3 ± 7.68 years; 21M/19F; BMI 28.9 ± 4.52 kg/m²) completed a 25 cm box StUD-F task. Surface electromyograph recorded bilateral activation of the vastus medialis (VM), vastus lateralis (VL), bicep femoris (BF), and semitendinosus (ST). Triplanar hip, knee, and ankle joint angles were estimated using inertial measurement units. StUD-F events (initial stance; step contact; ascent completion; descent preparation; step-down touchdown; and descent completion) were identified using custom algorithms. Pain was assessed using visual analogue scales and the Knee Injury and Osteoarthritis Outcome Score (KOOS). Limb differences were analysed for leading or trailing roles using paired samples t-tests or non-parametric equivalents; waveforms were visually inspected. Results: Distinct neuromuscular and kinematic asymmetries were observed when affected and contralateral limbs were compared within each role (leading/trailing). During step-up, the affected leading limb demonstrated higher quadriceps activation at initial stance (VM: p = 0.035; VL: p = 0.027) and reduced trailing-limb activation at step contact (VM: p = 0.015; VL: p = 0.018), with sagittal-plane ankle differences (p = 0.004). During step-down, when the affected limb initiated ascent, trailing limb activation was higher at descent completion (VL: p < 0.001; VM: p = 0.003; BF: p = 0.009), with coronal-plane hip deviations (p < 0.001). When the contralateral limb-initiated ascent, trailing-limb muscles activation differences (VM: p < 0.001; VL: p = 0.015; BF: p = 0.007) and ankle/coronal-plane asymmetries (p ≤ 0.049) persisted. Conclusions: The StUD-F task elicits altered strategies in KOA, including elevated quadriceps–hamstring co-activation and altered sagittal/coronal alignment, and habitual limb choice across ascent and descent. These adaptations may enhance stability and joint protection but could increase medial compartment loading. The findings support rehabilitation focused on dynamic control, alignment, and shock absorption. Full article

Background: Older adult falls during step negotiation result in higher injury rates compared to level ground falls. Previous research on discrete events during step negotiations may not capture important age-related changes. Curve analysis techniques enable assessment of an entire time series and may further advance the understanding of older adult falls during step negotiation. The purpose of the current study was to investigate lower extremity kinematics during transition step negotiation in older women with fall history compared to young women using statistical parametric mapping (SPM). Methods: 15 older female adults with a fall history and 15 young female adults participated in the study. Participants performed walking trials along a 5.5 m raised walkway, descended a 17 cm step and continued walking 3 m. Data was processed from lead limb toe-off prior to the step, through lead limb weight acceptance of the transition step. SPM was used to perform independent t-test analysis of the three-dimensional lower extremity time series. Results: The older faller group showed significantly decreased lead hip abduction (9–19% of step negotiation, mean difference: 3.74°, p = 0.045), increased lead knee flexion (65–80% of step negotiation, mean difference: 5.8°, p = 0.012), and increased trail limb hip adduction (91–100% of step negotiation, mean difference: 3.92°, p = 0.046). Conclusions: The older faller group showed altered hip joint angles in the frontal plane and knee joint angles in the sagittal plane during early swing and late weight acceptance phases, which may reflect compensatory strategies for reduced strength and/or balance. Curve analysis provides additional insight into age-related kinematic changes during step negotiation that may be related to older adult fall risk. Full article

For the first time, this paper introduces Nabla fractional calculus into the distributed Nash equilibrium (NE) seeking problem of aggregative games (AGs) with partial decision information in undirected communication networks, and proposes two novel fractional-order distributed algorithms. In the considered setting, each agent can access to only local information and collaboratively estimates the global aggregate through communication with its neighbors. Both algorithms adopt a backward-difference scheme followed by an implicit fractional-order gradient descent step. One updates local aggregate estimates via fractional-order dynamic tracking and the other uses fractional-order average dynamic consensus protocols. Under standard assumptions, convergence of both algorithms to the NE is rigorously proved using nabla fractional-order Lyapunov stability theory, achieving a Mittag-Leffler convergence rate. The feasibility of the developed schemes is verified via numerical experiments applied to a Nash-Cournot game and the coordination control of flexible robotic arms. Full article

The highly squinted mode, as an operational configuration of synthetic aperture radar (SAR), fulfills specific remote sensing demands. Under equivalent conditions, it necessitates a higher pulse repetition frequency (PRF) than the side-looking mode but produces inferior imaging quality, thereby constraining its widespread application. By applying the sparse SAR imaging method to highly squinted SAR systems, imaging quality can be enhanced while simultaneously reducing PRF requirements and expanding swath. Hyperparameters in sparse SAR imaging critically influence reconstruction quality and computational efficiency, making hyperparameter optimization (HPO) a persistent research focus. Inspired by HPO techniques in the deep unfolding network (DUN), we modified the iterative soft-thresholding algorithm (ISTA) employed in fast sparse SAR reconstruction based on approximate observation operators. Our adaptation enables adaptive regularization parameter tuning during iterations while accelerating convergence. To improve the robustness of this enhanced algorithm under realistic SAR echoes with noise, we integrated hypergradient descent (HD) to automatically adjust the ISTA step size after regularization parameter convergence, thereby mitigating overfitting. The proposed method, named Hyper-ISTA-GHD, adaptively selects regularization parameters and step sizes. It achieves high-precision, rapid imaging for highly squinted SAR. Owing to its training-free iterative minimization framework, this approach exhibits superior generalization capabilities compared to existing DUN methods and demonstrates broad applicability across diverse SAR imaging modes and scene characteristics. Simulations show that the hyperparameter selection and reconstruction results of the proposed method are almost consistent with the optimal values of traditional methods under different signal-to-noise ratios and sampling rates, but the time consumption is only one-tenth of that of traditional methods. Comparative experiments on the generalization performance with DUN show that the generalization performance of the proposed method is significantly better than DUN in extremely sparse scenarios. Full article

In this paper, we present a rapid gradient descent method for solving low-rank matrix recovery problems. Our method extends the conventional gradient descent framework by exploiting the problem’s unique features to develop an innovative fast gradient computation technique that lowers the computational cost of gradient evaluation. The introduced adaptive step size selection strategy not only eliminates the need for the heavy calculations usually involved in finding the descent direction but also guarantees a consistent decrease in the objective function at every iteration. Additionally, we offer a proof confirming the algorithm’s convergence. Numerical experiments are provided to show the efficiency of the proposed algorithm. Full article

Adaptive filtering remains a cornerstone of modern signal processing but faces fundamental challenges when confronted with rapidly changing or nonlinear environments. This work investigates the integration of deep learning into adaptive-filter architectures to enhance tracking capability and robustness in non-stationary conditions. After reviewing and analyzing classical algorithms—LMS, NLMS, RLS, and a variable step-size LMS (VSS-LMS)—their theoretical stability and mean-square error behavior are formalized under a slow-variation system model. Comprehensive simulations using drifting autoregressive (AR(2)) processes, piecewise-stationary FIR systems, and time-varying sinusoidal signals confirm the classical trade-off between performance and complexity: RLS achieves the lowest steady-state error, at a quadratic cost, whereas LMS remains computationally efficient with slower adaptation. A stabilized VSS-LMS algorithm is proposed to balance these extremes; the results show that it maintains numerical stability under abrupt parameter jumps while attaining steady-state MSEs that are comparable to RLS (approximately 3 × 10⁻²) and superior robustness to noise. These findings are validated by theoretical tracking-error bounds that are derived for bounded parameter drift. Building on this foundation, a deep-learning-driven adaptive filter is introduced, where the update rule is parameterized by a neural function, $U_{\theta }$, that generalizes the classical gradient descent. This approach offers a pathway toward adaptive filters that are capable of self-tuning and context-aware learning, aligning with emerging trends in AI-augmented system architectures and next-generation computing. Future work will focus on online learning and FPGA/ASIC implementations for real-time deployment. Full article

Data-driven quality control (QC) systems for the hot forming of steel parts increasingly rely on deep learning models deployed at the network edge, making multivariate sensor time series a critical asset for both local decisions and management information system (MIS) reporting. However, these models are vulnerable to adversarial perturbations and realistic signal disturbances, which can induce misclassification and distort key performance indicators (KPIs) such as first-pass yield (FPY), scrap-related losses, and latency service-level objectives (SLOs). To address this risk, this study introduces a Digital-Twin-Conditioned Diffusion Purification (DTCDP) framework that constrains latent diffusion-based denoising using process states from a lightweight digital twin of the hot-forming line. At each reverse-denoising step, the twin provides physics residuals that are converted into a scalar penalty, and the diffusion latent is updated with a guidance term. This directly bends the sampling trajectory toward reconstructions that adhere to process constraints while removing adversarial perturbations. DTCDP operates as an edge-side preprocessing module that purifies sensor sequences before they are consumed by existing long short-term memory (LSTM)-based QC models, while exposing purification metadata and physics-guidance diagnostics to the plant MIS. In a four-week production dataset comprising more than 40,000 bars, with white-box ℓ∞ attacks crafted on multivariate sensor time series using Fast Gradient Sign Method and Projected Gradient Descent at perturbation budgets of 1–3% of the physical range, combined with additional realistic disturbances, DTCDP improves the robust classification performance of an LSTM-based QC model from 61.0% to 81.5% robust accuracy, while keeping clean accuracy (≈93%) and FPY on clean data (≈97%) essentially unchanged. These results indicate that physics-aware, digital-twin-guided diffusion purification can enhance the adversarial robustness of edge QC in hot forming without compromising operational KPIs. Full article

Winter jujube harvesting is traditionally labor-intensive, yet declining labor availability and rising costs necessitate robotic automation to maintain agricultural competitiveness. Path planning for robotic arms in orchards faces challenges due to the unstructured, dynamic environment containing densely packed fruits and branches. To overcome the limitations of existing robotic path planning methods, this research proposes BMGA-RRT Connect (BVH-based Multilevel-step Gradient-descent Adaptive RRT), a novel algorithm integrating adaptive multilevel step-sizing, hierarchical Bounding Volume Hierarchy (BVH)-based collision detection, and gradient-descent path smoothing. Initially, an adaptive step-size strategy dynamically adjusts node expansions, optimizing efficiency and avoiding collisions; subsequently, a hierarchical BVH improves collision-detection speed, significantly reducing computational time; finally, gradient-descent smoothing enhances trajectory continuity and path quality. Comprehensive 2D and 3D simulation experiments, dynamic obstacle validations, and real-world winter jujube harvesting trials were conducted to assess algorithm performance. Results showed that BMGA-RRT Connect significantly reduced average computation time to 2.23 s (2D) and 7.12 s (3D), outperforming traditional algorithms in path quality, stability, and robustness. Specifically, BMGA-RRT Connect achieved 100% path planning success and 90% execution success in robotic harvesting tests. These findings demonstrate that BMGA-RRT Connect provides an efficient, stable, and reliable solution for robotic harvesting in complex, unstructured agricultural settings, offering substantial promise for practical deployment in precision agriculture. Full article

In constrained nonlinear optimization, we aim to achieve two goals: one is to minimize the objective function, and the other is to satisfy the constraints. A common way to balance these competing targets is to use penalty functions. Suppose that an algorithm generates a descent direction and produces a step that decreases the objective function value but increases the constraint violation—a phenomenon known as the Maratos effect. This leads to the rejection of the full step by the non-smooth penalty function; therefore, superlinear convergence is not preserved. This work leverages a piecewise convexity model to solve the optimal PMU placement. A quadratic objective function is minimized subject to a non-convex equality constraint within box constraints [0, 1] × [0, 1] ⊂ R². The initial non-convex region is reconsidered as a union of piecewise line segments. This decomposition enables algorithms to converge to a local optimum while preserving superlinear convergence near the solution. An analytical solution is presented using the Karush–Kuhn–Tucker conditions. First-and-second-order optimality conditions are applied to find the local minimum. We show how the Maratos effect is avoided by adopting the piecewise convexity without needing a non-smooth penalty function, second-order corrections or employing the watchdog methods. Simulations demonstrate that the algorithms partially search the space along the line segments—avoiding zig-zag trajectories—and reach (0, 1) or (1, 0), where both feasibility and optimality are satisfied at once. Full article

This paper investigates a UAV-assisted wireless communication system that integrates optical wireless communication (LiFi) with conventional RF links to enhance network capacity in crowd-gathering scenarios. While the unmanned aerial vehicle (UAV) serves as a flying base station providing downlink transmission to mobile ground users, the study places particular emphasis on the role of LiFi as a complementary physical layer technology within heterogeneous networks—an aspect closely connected to optical and photonics advancements. The proposed system is designed for environments such as theme parks and public events, where user groups move collectively toward points of interest (PoIs). To maintain quality of service (QoS) under dynamic mobility, we develop a joint optimization framework that simultaneously designs the UAV’s flight path and resource allocation over time. Given the problem’s non-convexity, a block coordinate descent (BCD) based approach is introduced, which decomposes the problem into power allocation and path planning subproblems. The power allocation step is solved using convex optimization techniques, while the path planning subproblem is handled via successive convex approximation (SCA). Simulation results demonstrate that the proposed algorithm achieves rapid convergence within 3–5 iterations while guaranteeing 100% heterogeneous QoS satisfaction, ultimately yielding nearly 15.00 bps/Hz system capacity enhancement over baseline approaches. These findings motivate the integration of coordinated three-dimensional trajectory planning for multi-UAV cooperation as a promising direction for further enhancement. Although LiFi is implemented in free-space optics rather than fiber-based sensing, this work highlights a relevant optical technology that may inspire future cross-domain applications, including those in optical sensing, where UAVs and reconfigurable optical links play a role. Full article

Gradient accumulation enables training large-scale deep learning models under GPU memory constraints by aggregating gradients across multiple microbatches before parameter updates. Standard gradient accumulation treats all microbatches uniformly through simple averaging, implicitly assuming that all stochastic gradient estimates are equally reliable. This assumption becomes problematic in non-convex optimization where gradient variance across microbatches is high, causing some gradient estimates to be noisy and less representative of the true descent direction. In this paper, FracGrad is proposed, a simple weighting scheme for gradient accumulation that biases toward recent microbatches via a power-law schedule derived from a discretized Riemann–Liouville integral. Unlike uniform summation, FracGrad reweights each microbatch gradient by $w_i\left(\alpha \right)={\displaystyle \frac{{(N-i+1)}^{\alpha }-{(N-i)}^{\alpha }}{{\sum }_{j=1}^N[{(N-j+1)}^{\alpha }-{(N-j)}^{\alpha }]}}$, controlled by $\alpha \in (0,1]$. When $\alpha =1$, standard accumulation is recovered. In experiments on mini-ImageNet with ResNet-18 using up to $N=32$ accumulation steps, the best FracGrad variant with $\alpha =0.1$ improves test accuracy from 16.99% to 31.35% at $N=16$. Paired t-tests yield $p\approx 2\times {10}^{-6}$. Full article

This paper presents a novel Ripple Evolution Optimizer (REO) that incorporates adaptive and diversified movement—a population-based metaheuristic that turns a coastal-dynamics metaphor into principled search operators. REO augments a JADE-style current-to-p-best/1 core with jDE self-adaptation and three complementary motions: (i) a rank-aware that pulls candidates toward the best, (ii) a time-increasing that aligns agents with an elite mean, and (iii) a scale-aware sinusoidal that lead solutions with a decaying envelope; rare Lévy-flight kicks enable long escapes. A reflection/clamp rule preserves step direction while enforcing bound feasibility. On the CEC2022 single-objective suite (12 functions spanning unimodal, rotated multimodal, hybrid, and composition categories), REO attains 10 wins and 2 ties, never ranking below first among 34 state-of-the-art compared optimizers, with rapid early descent and stable late refinement. Population-size studies reveal predictable robustness gains for larger N. On constrained engineering designs, REO achieves outperforming results on Welded Beam, Spring Design, Three-Bar Truss, Cantilever Stepped Beam, and 10-Bar Planar Truss. Altogether, REO couples adaptive guidance with diversified perturbations in a compact, transparent optimizer that is competitive on rugged benchmarks and transfers effectively to real engineering problems. Full article

Background: Glaucoma is the leading cause of global irreversible blindness, and it disproportionately affects people of African descent, in addition to having slightly higher prevalence rates in females. Glaucoma is a group of diseases that are characterized by progressive and irreversible damage to the optic nerve, leading to eventual blindness without proper treatment. There are a number of interventions available to treat glaucoma, including MIGS, of which usage has drastically increased due to its safety and efficacy. However, with minority populations, such as people of African descent, having the highest disease burden, it remains critical to evaluate the diversity of clinical trial populations that are used in the study of glaucoma treatments. The objective of this study is to compare the representation of Black and other ethnic minorities, as well as female participants, between cataract surgery (CS), minimally invasive glaucoma surgery (MIGS), and MIGS and cataract surgery (MACS) trials. Methods: This analysis consisted of publicly available data on MIGS, CS, and MACS clinical trials from 2005 to 2017, using ClinicalTrials.gov as well as prevalence data sourced from the CDC. Data reporting and synthesis adhered to PRISMA guidelines. This study focuses on sex rather than gender, as this is how data was reported on ClinicalTrials.gov. The primary outcome was the participation-to-prevalence ratio (PPR) of each clinical trial. A PPR between 0.8 and 1.2 represents adequate representation, while a PPR less than 0.8 or greater than 1.2 can signify under- or over-representation, respectively. Results: A total of 21 trials were included in this review, comprising 3330 clinical trial participants: 7 CS trials (N = 570), 13 MIGS trials (N = 1577), and 9 MACS trials (N = 1183). All of the clinical trials included data on sex, while only 14 reported race data and 7 reported ethnicity data. The overall PPR of female participants was 1.00, with CS, MIGS, and MACS clinical trials having PPRs of 0.99, 1.00, and 1.00, respectively. On the other hand, the overall PPR of Black participants was 0.44, with CS, MIGS, and MACS clinical trials having PPRs of 0.27, 0.62, and 0.22, respectively. Further analysis demonstrated that the PPR of Black participants in trials sponsored by medical device companies and medical centers or universities was 0.41 and 1.25, respectively. The study was registered with Prospero CRD420251152586. Conclusions: Cataract surgery, MIGS, and MIGS and cataract surgery clinical trials under-represent Black individuals and appropriately represent females. Due to the disproportionate amount of Black individuals impacted by glaucoma, this lack of representation raises concerns about the applicability of the clinical trials to these populations. Understanding clinical trial disparities in the representation of minority races is a key first step toward promoting advancements in diversity and equitable healthcare. Clinical trials in the future need to make a genuine effort to include minority groups to improve the generalizability of results. Full article

The gravity forward modeling algorithm is a compute-intensive method and is widely used in scientific computing, particularly in geophysics, to predict the impact of subsurface structures on surface gravity fields. Traditional implementations rely on CPUs, where performance gains are mainly achieved through algorithmic optimization. With the rise of domain-specific architectures, FPGA offers a promising platform for acceleration, but faces challenges such as limited programmability and the high cost of nonlinear function implementation. This work proposes an FPGA-based co-processor to accelerate gravity forward modeling. A RISC-V core is integrated with a custom instruction set targeting key computation steps. Tasks are dynamically scheduled and executed on eight fully pipeline processing units, achieving high parallelism while retaining programmability. To address nonlinear operations, we introduce a piecewise linear approximation method optimized via stochastic gradient descent (SGD), significantly reducing resource usage and latency. The design is implemented on the AMD UltraScale+ ZCU102 FPGA (Advanced Micro Devices, Inc. (AMD), Santa Clara, CA, USA) and evaluated across several forward modeling scenarios. At 250 MHz, the system achieves up to 179× speedup over an Intel Xeon 5218R CPU (Intel Corporation, Santa Clara, CA, USA) and improves energy efficiency by 2040×. To the best of our knowledge, this is the first FPGA-based gravity forward modeling accelerate design. Full article

(1) Background: Chronic ankle instability (CAI), a common outcome of ankle sprains, involves recurrent sprains, balance deficits, and gait impairments linked to both peripheral and central neuromuscular dysfunction. Dual-task (DT) demands further aggravate postural control, especially during stair descent, a major source of fall-related injuries. Yet the biomechanical mechanisms of stair-to-ground transition in CAI under dual-task conditions remain poorly understood. (2) Methods: Sixty individuals with CAI and age- and sex-matched controls performed stair-to-ground transitions under single- and dual-task conditions. Spatiotemporal gait parameters, center of pressure (COP) metrics, ankle inversion angle, and relative joint work contributions (Ankle%, Knee%, Hip%) were obtained using 3D motion capture, a force plate, and musculoskeletal modeling. Correlation and regression analyses assessed the relationships between ankle contributions, postural stability, and proximal joint compensations. (3) Results: Compared with the controls, the CAI group demonstrated marked control deficits during the single task (ST), characterized by reduced gait speed, increased step width, elevated mediolateral COP root mean square (COP-ml RMS), and abnormal ankle inversion and joint kinematics; these impairments were exacerbated under DT conditions. Individuals with CAI exhibited a significantly reduced ankle plantarflexion moment and energy contribution (Ankle%), accompanied by compensatory increases in knee and hip contributions. Regression analyses indicated that Ankle% significantly predicted COP-ml RMS and gait speed (GS), highlighting the pivotal role of ankle function in maintaining dynamic stability. Furthermore, CAI participants adopted a “posture-first” strategy under DT, with concurrent deterioration in gait and cognitive performance, reflecting strong reliance on attentional resources. (4) Conclusions: CAI involves global control deficits, including distal insufficiency, proximal compensation, and an inefficient energy distribution, which intensify under dual-task conditions. As the ankle is central to lower-limb kinetics, its dysfunction induces widespread instability. Rehabilitation should therefore target coordinated lower-limb training and progressive dual-task integration to improve motor control and dynamic stability. Full article