Search Results (1,661)

In disturbance scenarios, wheeled bipedal robots (WBRs) require effective control algorithms to restore balance. To address the trade-off between computational burden and control precision, and to enhance anti-disturbance capability, this paper proposes a soft-constrained Model Predictive Control (MPC) algorithm with optimized horizon parameters tailored to the hardware of the WBR. A cost function is designed, and the Dung Beetle Optimizer (DBO) is employed to optimize the MPC’s prediction and control horizons. An experimental platform is built, and impact and load disturbance experiments are conducted. The experimental results show that, under impact disturbances, the pitch angle and displacement overshoot with optimized MPC are reduced by 58.57% and 42.20%, respectively, compared to unoptimized LQR. Under load disturbances, the pitch angle and displacement overshoot are reduced by 17.09% and 15.53%, respectively, with both disturbances converging to the equilibrium position. Full article

Meissner’s corpuscles are essential mechanoreceptors that detect low-frequency vibrations. However, the internal fluid dynamic processes that convert directional mechanical stimuli into neural signals are not yet fully understood. This study aims to clarify the direction-specific sensing mechanism by analyzing internal fluid flow and shear stress distribution under different vibration modes. A biomimetic microfluidic platform was developed and coupled with a dynamic mesh computational fluid dynamics (CFD) model to simulate the response of the corpuscle to 20 Hz normal and tangential vibrations. The simulation results showed clear differences in fluid behavior. Normal vibration produced localized vortices and peak wall shear stress greater than 0.0054 Pa along the short axis. In contrast, tangential vibration generated stable laminar flow with a lower average shear stress of about 0.0012 Pa along the long axis. These results suggest that the internal structure of the Meissner corpuscle is important for converting mechanical inputs from different directions into specific fluid patterns. This study provides a physical foundation for understanding mechanotransduction and supports the design of biomimetic sensors with improved directional sensitivity for use in smart skin and soft robotic systems. Full article

Background: Lumbar disk herniation is a common cause of low back pain and radiculopathy, significantly impacting patients’ life quality and functional capacity. Magnetic Resonance Imaging (MRI) remains the gold standard for its assessment due to its superior soft tissue contrast and multiplanar imaging capabilities. However, recent advances in spectral computed tomography (CT), particularly dual-energy CT (DECT), have introduced new diagnostic opportunities, offering improved soft tissue characterization. Objective: To evaluate the diagnostic performance of DECT in detecting and grading lumbar disk herniations using dedicated color-coded fat maps. Materials and Methods: A total of 205 intervertebral levels from 41 consecutive patients with lumbar symptoms were prospectively analyzed. All patients underwent both DECT and MRI within 3 days. Three radiologists with varying years of experience independently assessed DECT images using color-coded reconstructions. A five-point grading score was attributed to each lumbar level: 1 = normal disk, 2 = bulging/protrusion, 3 = focal herniation, 4 = extruded herniation, and 5 = migrated fragment. The statistical analysis included Pearson’s correlation for score consistency, Cohen’s Kappa for interobserver agreement, generalized estimating equations for a cluster-robust analysis, and an ROC curve analysis. The DECT diagnostic accuracy was assessed in a dichotomized model (grades 1–2 = no herniation; 3–5 = herniation), using MRI as reference. Results: A strong correlation was observed between DECT and MRI scores across all readers (mean Pearson’s r = 0.826, p < 0.001). The average exact agreement between DECT and MRI was 79.4%, with the highest concordance at L1–L2 (86.7%) and L5–S1 (80.4%). The interobserver agreement was substantial (mean Cohen’s κ = 0.765), with a near-perfect agreement between the two most experienced readers (κ = 0.822). The intraclass correlation coefficient was 0.906 (95% CI: 0.893–0.918). The ROC analysis showed excellent performance (AUC range: 0.953–0.986). In the dichotomous model, DECT demonstrated a markedly higher sensitivity than conventional CT (95.1% vs. 57.2%), with a comparable specificity (DECT: 99.0%; CT: 96.5%) and improved overall accuracy (98.4% vs. 90.0%). Subgroup analyses by age and disk location revealed no statistically significant differences. Conclusions: The use of DECT dedicated color-coded fat map reconstructions showed high diagnostic performance in the assessment of lumbar disk herniations compared to MRI. These findings support the development of dedicated post-processing tools, facilitating the broader clinical adoption of spectral CT, especially in cases where MRI is contraindicated or less accessible. Full article

This work thoroughly investigated the compound 4-(2,5-Dimethoxyphenyl)-3,4-dihydrobenzo[g]chromene-2,5,10-trione (DMDCT) using molecular docking, quantum chemical analysis, and vibrational spectroscopy methodology. The medicinal chemistry group has been particularly interested in chromene and benzochromene derivatives due to their wide range of pharmacological actions, including anticancer, antibacterial, anti-inflammatory, antioxidant, antiviral, and neuroprotective capabilities. In this connection, DMDCT has been explored to evaluate its biological, electrical, and structural properties. DFT using the B3LYP functional and 6–31G basis was established to conduct theoretical computations with the Gaussian 09 program. The findings from these computations provide insight into the following topics: NBO interactions, optimal molecular geometry, Mulliken charge distribution, frontier molecular orbitals, and MEP. Second-order perturbation theory has been used to assess stabilization energies arising from donor–acceptor interactions. Furthermore, general features such as chemical hardness, softness, and electronegativity were studied. The results suggest that DMDCT has stable electronic configurations and biologically relevant active sites. This integrated experimental and theoretical study supports the potential of DMDCT as a practical scaffold for future therapeutic applications and contributes valuable information regarding its vibrational and electronic behavior. Full article

The recent demand for 8-bit floating-point (FP) formats is driven by their potential to accelerate domain-specific applications with intensive vector computations (e.g., machine learning, graphics, and data compression). This paper presents the design, implementation, and application of the software model of an 8-bit FP vector arithmetic operation set, compliant with the RISC-V vector instruction set architecture. The model has been developed as an extension of the SoftFloat library and integrated into the RISC-V reference instruction-level simulator Spike, providing the first open-source 8-bit SoftFloat extension for an instruction-set simulator. Based on the SoftFloat library templates for standard FP formats, the proposed extension implements the two widely used 8-bit formats E4M3 and E5M2 in both Open Compute Project (OCP) and IEEE 754 variants. In host-time micro-kernels, FP8 delivers +2–4% more elements per second versus FP32 (across vfadd/vfsub/vfmul) and ≈5% lower RSS; E4M3 and E5M2 perform similarly. Enabling FP8 in Spike increases the stripped binary by ~1.8% (mostly .text). The proposed extension was used to fully verify and correct errors in the vector FP unit design for the eProcessor European project, and continues to be used to verify other 8-bit FP unit implementations. Full article

Deploying deep learning (DL) models in real-world environments remains a major challenge, particularly under resource-constrained conditions where achieving both high accuracy and compact architectures is essential. While effective, Conventional pruning methods often suffer from high computational overhead, accuracy degradation, or disruption of the end-to-end training process, limiting their practicality for embedded and real-time applications. We present Dynamic Attention-Guided Pruning (DAGP), a Dynamic Attention-Guided Soft Channel Pruning framework that overcomes these limitations by embedding learnable, differentiable pruning masks directly within convolutional neural networks (CNNs). These masks act as implicit attention mechanisms, adaptively suppressing non-informative channels during training. A progressively scheduled L1 regularization, activated after a warm-up phase, enables gradual sparsity while preserving early learning capacity. Unlike prior methods, DAGP is retraining-free, introduces minimal architectural overhead, and supports optional hard pruning for deployment efficiency. Joint optimization of classification and sparsity objectives ensures stable convergence and task-adaptive channel selection. Experiments on CIFAR-10 (VGG16, ResNet56) and PlantVillage (custom CNN) achieve up to 98.82% FLOPs reduction with accuracy gains over baselines. Real-world validation on an enhanced PlantDoc dataset for agricultural monitoring achieves 60 ms inference with only 2.00 MB RAM on a Raspberry Pi 4, confirming efficiency under field conditions. These results illustrate DAGP’s potential to scale beyond agriculture to diverse edge-intelligent systems requiring lightweight, accurate, and deployable models. Full article

The Internet of Medical Things (IoMT) is the latest advancement in the Internet of Things (IoT). Researchers are increasingly drawn to its vast potential applications in secure healthcare systems. The growing use of internet-connected medical device sensors has significantly transformed healthcare, necessitating the development of robust methodologies to assess their integrity. As access to computer networks continues to expand, these sensors have become vulnerable to a wide range of security threats, thereby compromising their integrity. To prevent such lapses, it is essential to understand the complexities of the operational environment and to systematically identify technical vulnerabilities. This paper proposes a unified hesitant fuzzy-based healthcare system for assessing IoMT sensor integrity. The approach integrates the hesitant fuzzy Analytic Network Process (ANP) and the hesitant fuzzy Technique for Order Preference by Similarity to the Ideal Solution (TOPSIS). In this study, a hesitant fuzzy ANP is employed to construct a comprehensive network that illustrates the interrelationships among various integrity criteria. This network incorporates expert input and accounts for inherent uncertainties. The research also offers sensitivity analysis and comparative evaluations to show that the suggested method can analyse many medical device sensors. The unified hesitant fuzzy-based healthcare system presented here offers a systematic and valuable tool for informed decision-making in healthcare. It strengthens both the integrity and security of healthcare systems amid the rapidly evolving landscape of medical technology. Healthcare stakeholders and beyond can significantly benefit from adopting this integrated fuzzy-based approach as they navigate the challenges of modern healthcare. Full article

Power outages in non-faulted zones caused by system failures significantly reduce the reliability of distribution networks. To address this issue, this paper proposes a fault self-healing technique based on the integration of soft open points (SOPs) and network reconfiguration. A mathematical model for power restoration is developed. The model incorporates SOP operational constraints and the stochastic output of photovoltaic (PV) distributed generation. And this formulation enables the determination of the optimal network reconfiguration strategy and enhances the restoration capability. The study first analyzes the operational principles of SOPs and formulates corresponding constraints based on their voltage support and power flow regulation capabilities. The stochastic nature of PV power output is then modeled and integrated into the restoration model to enhance its practical applicability. This restoration model is further reformulated as a second-order cone programming (SOCP) problem to enable efficient computation of the optimal network configuration. The proposed method is simulated and validated in MATLAB R2019a. Results demonstrate that combining the SOP with the reconfiguration strategy achieves a 100% load restoration rate. This represents a significant improvement compared to traditional network reconfiguration methods. Furthermore, the second-order cone programming (SOCP) transformation ensures computational efficiency. The proposed approach effectively enhances both the fault recovery capability and operational reliability of distribution networks with high penetration of renewable energy. Full article

Background: Congenital talipes equinovarus (clubfoot) affects 1–2 per 1000 newborns worldwide. The Ponseti method, based on staged manipulations and casting, is the gold standard for correction. However, the biomechanical processes underlying these corrections remain poorly understood, as infants rarely undergo imaging. Computational modeling may offer a non-invasive approach to studying correction pathways and exploring novel applications, such as customized casts. Methods: We developed a proof-of-concept framework using iterative finite element analysis (iFEA) to approximate the surface-level geometric corrections targeted in Ponseti treatment. A 3D surface model of a training clubfoot foot was scanned, meshed, and deformed stepwise under applied computational loads. The model was assumed to be homogeneous and hyperelastic, and correction was quantified using Cavus, Adductus, Varus, Equinus, and Derotation angles. We also introduced a secondary adult leg 3D surface model to assess whether model simplification influences correction outcomes, by comparing a homogeneous soft tissue model with a non-homogeneous model incorporating bone structure. Results: In the training model, iFEA generated progressive deformations consistent with Ponseti correction, with mean angular deviations of ±3.2°. In the adult leg model, homogeneous and non-homogeneous versions produced comparable correction geometries, differing by <2° in outcomes. The homogeneous model required less computation, supporting its use for feasibility testing. Applied loads were computational drivers, not physiological forces. Conclusions: This feasibility study shows that iFEA can reproduce surface-level geometric changes consistent with Ponseti correction, independent of model homogeneity. While not replicating clinical biomechanics, this framework lays the groundwork for future work that incorporates clinician-applied forces, pediatric tissue properties, and patient-specific geometries, with potential applications in customized 3D-printed casts. Full article

Real-time distribution system state estimation is hampered by limited observability, frequent topology changes, and measurement errors. Neural networks can capture the nonlinear characteristics of power-grid operation through a data-driven approach that possesses important theoretical value and is promising for engineering applications. In that context, we develop a deep learning framework that leverages General Attributed Multiplex Heterogeneous Network Embedding to explicitly encode the multiplex, heterogeneous structure of distribution networks and to support inductive learning that adapts to dynamic topology. A cross-modal attention mechanism further models fine-grained interactions between input measurements and node/edge attributes, enabling the capture of nonlinear correlations essential for accurate state estimation. To ensure physical feasibility, soft power-flow residuals are incorporated into training as a physics-constrained regularization, guiding predictions toward consistency with grid operation. Extensive studies on IEEE/CIGRE 14-, 70-, and 179-bus systems show that the proposed method surpasses conventional weighted least squares and representative neural baselines in accuracy, convergence speed, and computational efficiency while exhibiting strong robustness to measurement noise and topological uncertainty. Full article

Blind image deblurring, which requires simultaneous estimation of the latent image and blur kernel, constitutes a classic ill-posed problem. To address this, priors based on $L_2$, $L_1$, and $L_p$ regularizations have been widely adopted. Based on this foundation and combining successful experiences of previous work, this paper introduces $L_N$ regularization, a novel nonlinear sparse regularization combining the $L_p$ and $L_{\mathrm{\infty }}$ norms via nonlinear coupling. Statistical probability analysis demonstrates that $L_N$ regularization achieves stronger sparsity than traditional regularizations like $L_2$, $L_1$, and $L_p$ regularizations. Furthermore, building upon the $L_N$ regularization, we propose a novel nonlinear sparse model for blind image deblurring. To optimize the proposed $L_N$ regularization, we introduce an Adaptive Generalized Soft-Thresholding (AGST) algorithm and further develop an efficient optimization strategy by integrating AGST with the Half-Quadratic Splitting (HQS) strategy. Extensive experiments conducted on synthetic datasets and real-world images demonstrate that the proposed nonlinear sparse model achieves superior deblurring performance while maintaining completive computational efficiency. Full article

Soft biometric prediction—including age, gender, and ethnicity—is critical in surveillance applications, yet often suffers from performance degradation as the subject-to-camera distance increases. This study hypothesizes that embedding distance-awareness into the training process can mitigate such degradation and enhance model generalization across varying visual conditions. We propose a distance-adaptive, multi-task deep learning framework built upon EfficientNetB3, augmented with task-specific heads and trained progressively across four distance intervals (4 m to 10 m). A weighted composite loss function is employed to balance classification and regression objectives. The model is evaluated on a hybrid dataset combining the Front-View Gait (FVG) and MMV annotated pedestrian datasets, totaling over 19,000 samples. Experimental results demonstrate that the framework achieves up to 95% gender classification accuracy at 4 m and retains 85% accuracy at 10 m. Ethnicity prediction maintains an accuracy above 65%, while age estimation achieves a mean absolute error (MAE) ranging from 1.1 to 1.5 years. These findings validate the model’s robustness across distances and its superiority over conventional static learning approaches. Despite challenges such as computational overhead and annotation demands, the proposed approach offers a scalable and real-time-capable solution for distance-resilient biometric systems. Full article

In this research, we explore the computational and experimental optimization of compliant constant-torque mechanisms (CTMs) fabricated via injection molding using polymeric materials. We investigate how geometric variations influence the torsional strength of CTMs through numerical simulation, experimental validation, and artificial neural network (ANN) modeling. Four different geometries with the same overall dimensions were designed and analyzed to quantify their mechanical performance. The results reveal that the geometric configuration significantly affected the torsional behavior of the CTMs, with circular cross-sections demonstrating superior strength. Moreover, the ANN model exhibited a high prediction accuracy and minimal relative errors, closely aligning with the experimental outcomes. Despite this, discrepancies between our numerical and experimental results suggest that further refinements in material modeling and manufacturing processes are warranted. In this paper, we emphasize the importance of integrating computational (CAE), artificial neural networks (ANNs) and experimental techniques for optimizing polymer-based CTMs. CAE simulations for Model 4 showed a constant-torque section from 23–44 degrees with 0.142 N·m torque, while experimental and ANN results indicated a longer range (20–45/46 degrees) with higher torque values (0.164 N·m). Experimental and ANN predictions for Model 4 showed an approximate 97% similarity. While these findings represent a foundational step, the characteristics of polymer CTMs suggest potential relevance for advancing applications in precision engineering, biomedical devices, and soft robotics, pending further application-specific validation. Full article

One of the most common malocclusions encountered in everyday practice by orthodontists is skeletal Class II malocclusion, namely a protrusion of the maxilla, a retrusion of the mandible or a combination of both. To correct it, many clinicians use functional devices that guide the mandible into a more forward position. This stimulates bone growth, correcting the skeletal discrepancy. Controversy exists as to whether these appliances accelerate the growth rate, helping the mandible reach its final size earlier, or whether the growth of the mandible is observed as a positive response to the stimuli. This study examined whether the protrusion of the mandible in rats accelerates the growth rate or increases the overall growth of the mandible in the long run. Relapse was also assessed by removing the appliance prior to the end of the experiment. Seventy-two four-week-old Wistar rats were used. The treatment group, which consisted of 36 rats, had a device fitted on their upper incisors that led to a protrusion of their mandible. The device, a bite-jumping appliance, consisted of an iron-cast inclined plane and was fitted for 24 h a day, inducing a 3.5 mm anterior protrusion and 3 mm inferior displacement of the mandible. The control group consisted of 36 rats that were fed the same soft diet as the treatment group. Both groups were divided into three subgroups. The first was sacrificed 30 days after the onset of the experiment, the second at 60 days, and the last subgroup had the appliance removed for 30 days and was sacrificed 90 days after the onset of the experiment. At the beginning of the experiment, as well as at each time interval prior to the sacrifice of the animals, the appliances were removed, and cone beam-computed tomography was performed on every animal. Linear measurements were made on each 3D scan, measuring the growth of the mandible. Measurements of mandibular growth were higher compared to the control group. For instance, Gonion-Menton was 1.18 mm higher on month 2 compared to month 1 in the control group, whereas the same measurement marked a 1.82 mm difference in the experimental group. Condylion–Menton on the same intervals marked a 0.84 mm difference in the control, whereas a 1.35 mm difference was noted in the experimental group. Given the results, true mandibular growth is achieved using functional appliances for Class II malocclusion correction in rats. Full article

Background and Clinical Significance: Tooth impaction is a developmental anomaly characterized by the inability of a tooth to emerge into its predetermined anatomical position within the oral cavity during the normal eruption period. Impaction of the mandibular second premolar is an uncommon condition and poses a heightened risk of neurosensory injury when the tooth is adjacent to the mental foramen. Early diagnosis and precise planning are therefore essential. Case Presentation: This case report presents a rare instance of an asymptomatic impacted mandibular second premolar located in close proximity to the mental foramen in a 44-year-old female patient. The impaction was discovered incidentally on an orthopantomogram, and Cone-Beam Computed Tomography (CBCT) confirmed intimate contact between the root of the impacted second premolar and the mental nerve. Surgical removal was performed under local anesthesia via a conservative triangular flap and a corticotomy window. Platelet-Rich Fibrin (PRF) generated from autologous blood was placed in the socket to foster healing. The proximity of the mental foramen dictated minimal bone removal and atraumatic luxation to avoid nerve stretch or compression. PRF was selected as an effective biomaterial shown to accelerate soft tissue healing and moderate postoperative discomfort, potentially reducing the likelihood of neurosensory disturbance. The socket presented satisfactory healing, and neurosensory function was normal at the first week follow-up and remained normal at 7 months postoperatively (longest follow-up), and no complications were reported by the patient. Conclusions: CBCT-guided planning, meticulous surgical techniques, and adjunctive PRF allowed for safe extraction without post-operative paraesthesia. Timely identification of such rare impactions broadens treatment options and minimizes complications. Full article