Search Results (6,005)

Background/Objectives: Magnetically responsive biomaterials have emerged as promising platforms for tissue engineering as they allow remote stimulation of cells and improved control over tissue regeneration. Although gelatin–alginate hydrogels incorporating iron oxide nanoparticles have been reported, the use of U-type hexaferrite particles, particularly Al³⁺-substituted compositions, remains largely unexplored. This study aimed to develop and characterize extrusion-based 3D-printed gelatin–alginate scaffolds containing U-type hexaferrite particles and to evaluate the influence of particle composition and loading on the rheological, physicochemical and magnetic properties of the resulting biomaterials, as well as their in vitro compatibility with macrophages. Methods: Gelatin–alginate inks containing two U-type hexaferrite compositions (Ba₄Co₂Fe₃₆−_xAl_xO₆₀; x = 0.0 and x = 1.0) at two particle loadings (20 and 200 mg) were prepared and processed by extrusion-based 3D printing. The scaffolds were characterized by rheological analysis, SEM-EDS, FTIR, swelling measurements, magnetic responsiveness, and in vitro biological evaluation using RAW264.7 macrophages. Results: The inks exhibited suitable shear-thinning behavior and viscoelastic properties for extrusion-based printing. Increasing particle loading enhanced the thermal resistance of the network, whereas Al³⁺ substitution modified the viscoelastic response of the polymeric network. The 3D scaffolds successfully responded to an external magnetic field and SEM-EDS confirmed the homogeneous incorporation of hexaferrite particles. The magnetic scaffolds did not compromise macrophage metabolic activity. Conclusions: U-type hexaferrite particles provide an effective strategy for producing 3D-printed magnetically responsive scaffolds with tunable rheological properties without inducing an inflammatory response. The combined modulation of particle composition and loading represents a versatile approach for designing multifunctional inks with potential applications in tissue engineering. Full article

Visual–Inertial Odometry (VIO) has become a key technology for motion estimation in robotics and autonomous systems. However, deploying accurate VIO pipelines on embedded platforms remains challenging due to the trade-off between estimation accuracy, real-time performance, and energy consumption. This paper presents a hardware-aware Feature-PLPD VIO framework that integrates a point-and-line visual front-end with a loosely coupled Error-State Extended Kalman Filter (ESEKF) back-end. The proposed approach follows Algorithm Architecture Adequacy (A3) principles to preserve estimation robustness while limiting computational and memory requirements. To investigate its scalability across the considered resource constraints, two end-to-end embedded implementations are developed: a performance-oriented stereo VIO system on a GPU-based platform using GPU-aware software design, and a frugality-oriented low-cost RGB-D-assisted monocular VIO system on an FPGA-based architecture using hardware–software co-design with depth scale correction. The stereo GPU-based implementation is evaluated offline in both outdoor and indoor environments and is additionally validated through real-time on-the-fly deployment on a Scout Mini robot, whereas the FPGA-based implementation is evaluated offline using the indoor VICON dataset. Experimental results demonstrate meter-level trajectory accuracy and real-time performance under strict resource constraints. Averaged over four KITTI sequences, the proposed ESEKF-based fusion reduces the translation and rotation ATE by approximately 25% and 18%, respectively, compared with the corresponding VO-only configuration, while a 10% reduction in translation ATE is achieved in the indoor VICON environment, resulting in a normalized ATE of $5.09\%$ over the $23.77$ m trajectory. Both embedded implementations sustain around 20 fps, and the architectural evaluation highlights complementary accuracy–runtime–energy trade-offs between the GPU- and FPGA-based solutions. These results demonstrate the feasibility of scaling the proposed embedded VIO framework toward resource- and energy-constrained robotic applications under the investigated experimental and hardware configurations. Full article

This study proposes a lightweight domain-assisted deep learning framework for binary human presence detection using ultra-wideband (UWB) radar. The proposed methodology processes raw UWB radar signals through statistically screened, physics-grounded signal features including Fast Fourier Transform (FFT)-based frequency-domain statistics and Hilbert Transform (HT)-derived envelope statistics which are selected via a per-subject Cohen’s d screening step and stacked as auxiliary input channels alongside the raw signal for a lightweight two-dimensional convolutional neural network (2D-CNN). A cross-subject evaluation protocol (train-on-one-subject, test-on-the-other) is adopted to assess generalization across individuals rather than relying on a pooled, sample-level split. Among the candidate features, a Frequency Standard Deviation (FSTD) is shown to match or exceed the performance of every multi-feature combination tested, indicating that targeted feature selection is more consequential than input fusion for this task. To further improve deployment efficiency, post-training INT8 quantization is applied, reducing the model to approximately 23 KB while preserving classification performance for quantization-robust configurations. Hardware-in-the-loop benchmarking on the STEdgeAI platform indicates on-device inference times ranging from approximately 0.88 ms on AI-enabled STM32N6 hardware to 117–130 ms on STM32H7-class microcontrollers; these figures reflect model inference only and exclude radar acquisition and preprocessing time. Experiments are conducted on a two-subject (one male, one female) indoor dataset; the reported cross-subject results are presented as a relative comparison across feature and quantization configurations rather than as an estimate of population-level generalization. The findings nonetheless illustrate the feasibility of combining principled feature selection with quantization-aware, hardware-validated deployment on embedded artificial intelligence (AI) platforms. Full article

Reliable local three-dimensional (3D) wind estimates are important for fixed-wing UAV flight under wind disturbances, but low-cost platforms lack direct 3D flow sensing. We propose PIRNN-AKF, which combines a physics-informed gated recurrent unit (PI-GRU) with a measurement-noise-adaptive Kalman smoother. PI-GRU embeds the wind triangle and attitude rotations in an airspeed-closure loss and predicts covariance scales for adaptive fusion. In PX4/JSBSim simulations with Dryden turbulence, five-seed PI-GRU attains a 3D RMSE of $0.208\pm 0.003$ m/s and a direction MAE of 3.22° on Test-ID. PIRNN-AKF attains $0.544\pm 0.007$ m/s and 1.97° on Test-OOD, reducing RMSE by 37.4% relative to Vanilla GRU; KalmanNet yields lower OOD RMSE but weaker ID accuracy and direction estimation. Session-aware evaluation shows that the AKF reduces OOD jitter by 24.8% while preserving RMSE. In ten frozen HITL sessions of the final 41-input, 100-step model, ID/OOD RMSEs are $0.569\pm 0.054$/$0.832\pm 0.318$ m/s, the companion-side processing $p_{95}$ is $20.4\pm 0.5$ ms, and the achieved loop rate is $83.2\pm 1.6$ Hz with $5.6\pm 0.7\%$ deadline misses. These results support 50 Hz feasibility under simulation truth; real-flight accuracy and closed-loop benefits remain untested. Full article

This study presents a second-order adaptive fuzzy logic controller (SO-AFLC) to enhance the dynamic performance of permanent magnet synchronous generator (PMSG)-based wind energy conversion systems (WECSs). The proposed controller simultaneously performs maximum power point tracking (MPPT), DC-link voltage regulation, and reactive power control and minimizes speed-tracking errors. Its performance is evaluated under step-changing wind conditions and measured wind speed data from Ras Gharib, Gulf of Suez, Egypt. A comprehensive comparison with conventional proportional-integral (PI) and adaptive fuzzy logic controller (AFLC) methods is conducted using MATLAB/Simulink. Compared with the AFLC and PI controllers, the proposed SO-AFLC achieves superior rotor speed tracking performance, with improvements of 36.98% and 53.26%, respectively. The proposed controller is further validated experimentally using a dSPACE DS1104 real-time platform. Additionally, an overall Integral Absolute Error (IAE)-based wind performance index is introduced to enable an objective comparison of the investigated controllers under identical operating conditions. SO-AFLC decreases the average of IAEs by 9.15% and 21.36% compared with the AFLC and PI controllers, respectively. Both simulation and experimental results demonstrate that the SO-AFLC provides faster transient response, higher tracking accuracy, and more reliable energy conversion, making it a promising solution for improving the grid integration of PMSG-based WECSs. Full article

Background/Objectives: Three-dimensional (3D) heads-up visualization systems are increasingly used in vitreoretinal surgery and may offer advantages in intraoperative visualization, ergonomics, surgical education, and reduced light exposure. This study aimed to compare morphofunctional outcomes, intraoperative parameters, postoperative complications, and surgical team satisfaction between 3D heads-up visualization and standard operating microscopy in idiopathic epiretinal membrane (ERM) surgery. Methods: In this prospective randomized study, 25 eyes of 25 patients with idiopathic ERM were assigned to surgery using either a 3D heads-up visualization system (3D group, n = 13) or a standard operating microscope (SOM group, n = 12). BCVA, CMT, mfERG parameters, intraoperative variables, postoperative OCT-based macular complications, and satisfaction scores were evaluated. No formal sample-size calculation was performed; therefore, all analyses should be interpreted as exploratory and hypothesis-generating. Results: Postoperative BCVA and CMT were not significantly different between the two groups at both 1 and 3 months, suggesting that no major between-group difference was detected in final short-term visual and anatomical outcomes compared with standard operating microscopy in macular surgery. However, the 3D group had significantly worse preoperative BCVA, which may have inflated unadjusted change-score estimates. Therefore, the greater ΔBCVA observed in the 3D group should not be interpreted as evidence of superior efficacy. Conclusions: In idiopathic ERM surgery, 3D heads-up visualization showed no statistically significant difference in final anatomical and visual outcomes compared with standard operating microscopy, while allowing significantly lower endoillumination settings during peeling. The findings support its feasibility as an alternative visualization platform; however, they do not establish superiority, equivalence, or safety because of the exploratory design, small sample size, short follow-up, and baseline imbalances. Full article

Temporal lobe epilepsy (TLE) is increasingly conceptualized not as an isolated hippocampal lesion, but as a complex, multiscale limbic network connectomic disorder. Despite advances in pharmacological management, over 30% of patients experience drug-resistant epilepsy, underscoring the urgent need to shift from symptomatic seizure control to mechanism-directed disease modification. This review comprehensively synthesizes the pathophysiological architecture of epileptogenesis in TLE, spanning mitochondrial bioenergetic alterations, chronic neuroinflammation, synaptic reorganization, and ionic plasticity resulting from ion-channel dysregulation. We evaluate diagnostic advancements, highlighting how invasive stereo-EEG disambiguates pathological high-frequency oscillations from physiological ripples, how structural HARNESS-MRI maps anatomical substrates, and how AI-driven algorithms analyze ultra-long-term EEG streams for continuous seizure forecasting. Additionally, peripheral biofluid proteins and microRNAs provide noninvasive windows into active neuroinflammation and network remodeling, serving as valuable tools for longitudinal disease monitoring rather than primary screening. Therapeutically, the field is evolving beyond empirical antiseizure medications toward mechanism-based rational drug design and precision interventions. Dual-mechanism agents enhance seizure freedom, while antisense oligonucleotides, microRNA antagomirs, and cation-chloride cotransporter modulators target underlying genetic and biophysical drivers. Minimally invasive ablation, AI-guided closed-loop neuromodulation, targeted anti-inflammatory biologics, and patient-derived 3D cerebral organoid platforms further expand the translational frontier. Ultimately, bridging these experimental modalities through prospective clinical validation may provide a viable path toward interrupting epileptogenesis and realizing true disease modification in human TLE. Full article

Optimizing acute performance during strength training and competition has become an important objective for coaches and athletes. This study aimed to analyze the acute effect of supramaximal isometric holds (SMHs) on squat performance integrating barbell kinematics, force production, and postural stability. Fourteen resistance-trained subjects completed six experimental sessions encompassing squats at 60% and 80% of one-repetition maximum (1RM) following a 10 s SMH at 110% of 1RM with either 1 min or 5 min recovery, or in a control condition (CON) with no SMH. Barbell kinematics were assessed using a linear position transducer, as were vertical ground reaction force, inter-limb asymmetries, and center-of-pressure metrics via dual-force platforms. An SMH with 1 min recovery produced significant increases in mean propulsive velocity at both intensities compared to CON (+0.05 m·s⁻¹; d ≥ 0.94), a higher rate of force development at 80% of 1RM (p < 0.01), and non-significantly increased peak concentric force at 80% of 1RM (p = 0.056). An SMH with 5 min recovery had little to no effect. No significant differences were found in inter-limb force asymmetries or center-of-pressure metrics. A 10 s SMH coupled with a 1 min recovery period is a practical and costless strategy to enhance squat performance in strength-trained individuals through increased barbell velocity, force production, and unchanged postural stability. Full article

Sustained postoperative anti-inflammatory therapy following intraocular surgery remains an unmet clinical need, as conventional topical corticosteroid eye drops suffer from poor bioavailability, patient non-compliance, and an inability to maintain therapeutic drug levels over the multi-week inflammatory resolution period. Here we report the development and characterization of alkaline phosphatase (ALP)-responsive enzyme-instructed self-assembling (EISA) peptide hydrogels based on the ultrashort amphiphilic sequence NapFFKY(p)-OH, covalently conjugated with dexamethasone (DEX) via an ester-cleavable succinyl linker, in both L- and D-enantiomeric forms, as injectable platforms for subconjunctival drug delivery. Both conjugates, L-NapFFK(DEX)Y(p)-OH and D-NapFFK(DEX)Y(p)-OH, were synthesized with chemical purity (87.3% and 89.8%, respectively) and theoretical DEX loading of 29.58% w/w. ALP-triggered gelation was confirmed for both enantiomeric gels at 2% w/v, with rheological characterization of the platform establishing a storage modulus plateau of approximately 294 Pa, a G′/G″ ratio of ~10:1, and gel failure at approximately 578% shear strain. Transmission electron microscopy revealed stereochemistry-dependent differences in nanofiber network architecture, with the D-enantiomer forming a comparatively looser, more porous fibrillar matrix than its L-counterpart. In vitro DEX release over 28 days demonstrated that covalent conjugation eliminates the burst release associated with physical drug entrapment (41.1% for the L physical mixture versus 14.2% and 8.3% for L- and D-conjugated systems, respectively, at 2% w/v), while D-amino acid stereochemistry provides an additional level of sustained release restraint. Proteinase K stability studies confirmed the exceptional proteolytic resistance of the D-enantiomer, retaining 92.6 ± 2.2% of intact peptide at 6 h compared with 25.3 ± 0.8% for the L-enantiomer. Ex vivo scleral permeation studies demonstrated a trend towards greater cumulative DEX permeation (10.6 vs. 6.7 µg) and tissue retention (5.6 vs. 3.2 µg) for the D vs. L-formulation over 24 h. Both formulations exhibited injection forces within acceptable limits for a 30-gauge needle for ophthalmic administration, with the D-enantiomer requiring lower force than its L-counterpart. Cell viability of Dex-conjugated peptide sequence demonstrated non toxic in ARPE-19 human retinal pigment epithelial cells across all conditions tested, confirming biocompatibility. These findings establish the backbone stereochemistry independently controls network mechanics, proteolytic stability, and drug release rate in ALP-responsive EISA hydrogels. Cumulative release over 28 days remains below that required to cover a 4–6-week postoperative course, and this work is therefore presented as a mechanistic proof-of-concept defining design principles for the platform rather than a clinically optimized formulation. Drug loading, peptide sequence, and linker chemistry are identified as the levers through which release rate may be matched to intended clinical interval. Full article

This study investigates the effect of reheating on previously deposited layers in material extrusion (MEX) 3D printing, commonly known as Fused Deposition Modeling (FDM), considering the combined effects of deposition sequence (unidirectional and bidirectional) and key process parameters, including printing temperature, printing speed, and layer height. Two temperature measurement methods were used to record changes during printing, and the results from local and global approaches were compared. The temperature of the interface between the first deposited layer and the build platform was measured with thermocouples. The findings show that the reheating effect at the first layer fades with increasing build layer, with no further increase in heating profile after the eighth layer, regardless of the process parameters investigated. The measured minimum temperatures at the investigated location remained above the glass transition temperature, T_g, which is critical for bonding and polymer chain rearrangement to continue taking place. Printing temperature was found to be the strongest observed factor influencing reheating, as it provides the heat for conduction into previous layers. Deposition sequence also played an important role, with bidirectional deposition leading to higher reheating and 25–50% more layers above the crystallization temperature, T_c. Overall, the results confirm that deposited roads undergo cyclic heating, and understanding the influence of process parameters on reheating is important for interpreting bond formation in MEX 3D printing. Full article

Electromagnetic vibration and noise issues of converter transformers under harmonic operating conditions have become increasingly prominent. To enable a systematic comparison of different filtering strategies, this paper establishes a multiphysical field coupling model incorporating electromagnetic, structural, and acoustic fields, and systematically investigates four filtering configurations: grid-side LC filtering, valve-side LC filtering, inductive filtering, and valve-side capacitive filtering through theoretical analysis, finite-element simulation, and experimental verification on a 12-pulse laboratory HVDC prototype platform. The simulation results show that the maximum vibration acceleration decreases from 3.06 m/s² under the no-filter condition to 1.02 m/s² under valve-side capacitive filtering. Experimental measurements further confirm the suppression effect, with the vibration acceleration at Point 8 decreasing from 2.28 m/s² to 0.42 m/s², corresponding to a reduction of approximately 82%, while the average A-weighted sound pressure level decreases from 66.1 dB to 58.1 dB. These results indicate that vibration and noise suppression cannot be evaluated solely by total harmonic distortion; the attenuation and distribution of individual harmonic components should also be considered. The findings provide a theoretical and engineering basis for the vibration and noise reduction design of converter transformers. Full article

Automated manhole cover detection, a key task in urban infrastructure inspection, is hindered by modality-specific limitations: conventional monocular image-based detectors do not directly quantify cover-to-road elevation differences, LiDAR-based methods suffer from sparse sampling that limits recall for small, distant targets, and existing fusion approaches suffer from cross-modal spatial misalignment and lack 3D defect quantification. This paper proposes SDFA-Net, an end-to-end fisheye image and LiDAR point cloud fusion framework for manhole cover detection and diagnosis in wearable mobile mapping scenarios. The semi-dense depth-supervised view transformation (SDS-VT) module constructs semi-dense depth ground truth from multi-frame accumulated point clouds to explicitly supervise image-to-bird’s-eye-view (BEV) feature projection, addressing cross-modal spatial registration under wide-angle fisheye distortion. The fidelity-adaptive fusion (FA-Fusion) module builds a physical fidelity map by combining point-cloud projection density with depth-prediction entropy, then performs quality-aware adaptive fusion via cascaded channel-spatial attention. The geometry-aware decoupled detection head (Geo-Head) adopts an anchor-free decoupled architecture to jointly predict 2D localization, physical dimensions, and millimeter-level relative elevation in a single forward pass. A multimodal dataset of 3500 spatiotemporally aligned frames is constructed using the NavVis VLX wearable platform, covering five defect categories, with Random Sample Consensus (RANSAC)-derived relative-elevation reference labels independently validated by digital leveling. Experiments show that SDFA-Net achieves 92.9% mAP@0.5 and 63.5% mAP@0.5:0.95 with 18.5 M parameters, improving over the best single-modal baseline by 4.6, 11.0, and 8.7 percentage points in mAP@0.5, mAP@0.5:0.95, and Recall, respectively. Compared with BEVFusion, SDFA-Net achieves higher detection accuracy while maintaining a smaller overall model size and an inference speed of 35 FPS. Full article

We theoretically investigate the plasmonic analog of the relativistic Klein tunneling processes in systems of two coupled graphene waveguide arrays with spatially varying chemical potentials. By introducing an offset in the chemical potentials of the graphene sheets, a potential step is generated in this platform. Numerical simulations indicate that in the graphene array with $\lambda$ = 10 $\mathsf{\mu }$m, d = 70 nm, and ${\mu }_c\sim$ 0.15 eV, the surface plasmon polaritons exhibit non-unity transmission (up to 50%) across the potential step at the interface that depends on both angle and potential, which is analogous to the Klein tunneling effect. Analytical calculations of the transmission rate of Klein tunneling in these structures have been derived and are consistent with the simulations. This work provides a robust platform for controlling light propagation at deep-subwavelength scales and emulating relativistic quantum phenomena in photonic systems. Full article

The increasing proliferation of unmanned aerial vehicles (UAVs) has created a growing need for passive sensing techniques capable of operating in dynamic environments. This paper presents a passive drone-to-drone approach in which an observer UAV equipped with a microphone array captures the sound emitted by a target drone during flight. A statistical harmonic signature of the observer’s own platform is first constructed from repeated multichannel recordings and used for selective self-noise (ego-noise) suppression; the preserved tonal components of the target are then processed by multichannel time-delay estimation to obtain its azimuth relative to the array. Experiments on the AIRA-UAS dataset show consistent suppression of the selected observer-UAV harmonics (10.49–11.84 dB) with limited modification of the non-targeted broadband content, and produce azimuth sequences that are temporally stable and consistent with the expected motion of the target. As the dataset does not provide bearing ground truth synchronized with the audio, the study is presented as a feasibility assessment of self-noise suppression and bearing-trend estimation rather than of absolute localization accuracy, highlighting the potential of airborne acoustic sensing as a low-cost, passive component for UAV awareness applications. Full article

Background/Objectives: Alzheimer’s disease (AD) is preceded by clinically defined syndromes of subjective cognitive decline (SCD) and mild cognitive impairment (MCI), which are etiologically heterogeneous and belong to the AD continuum only when amyloid and tau pathology is biologically confirmed. Electroencephalography (EEG) microstates, brief periods of quasi-stable large-scale neural synchrony, have emerged as candidate markers of resting-state network disruption. This systematic review evaluated EEG microstate differences across clinically defined SCD, MCI, and AD dementia samples and the extent to which these have been related to amyloid/tau/neurodegeneration (AT(N)) biomarkers. Methods: PubMed/MEDLINE and EMBASE were searched (January 1990 to June 2026) on 30 June 2026 following preferred reporting items for systematic reviews and meta-analyses (PRISMA) 2020 guidelines, with an expanded search of Scopus, IEEE Xplore, Web of Science, PROSPERO, ClinicalTrials.gov, and WHO International Clinical Trials Registry Platform (searched 5 August 2026). Twenty-six primary studies met the eligibility criteria and were synthesized narratively. Two recently published meta-analyses identified by the same search were not counted among the included studies; they were appraised using AMSTAR 2 and used only to benchmark the primary-study synthesis, avoiding double counting of overlapping datasets. Methodological quality was appraised using the Newcastle–Ottawa Scale adapted for cross-sectional studies. Results: A graded pattern of microstate differences was identified across diagnostic groups. Microstate A duration and coverage were significantly increased at both the MCI and AD dementia stages. Microstate D duration, occurrence, and coverage were reduced at the MCI stage in the pooled contextual estimates, although only two independent primary samples contributed a microstate D contrast at this stage, whereas only microstate D occurrence was significantly reduced at the AD dementia stage. Microstate C occurrence was significantly reduced at the AD dementia stage in the pooled contextual estimate, but no individual primary sample showed a significant reduction and two of six reported an increase, so the pooled and primary-study evidence diverge for this parameter. Reductions in microstate C at the SCD stage were reported by a single primary study and were associated with cerebrospinal fluid amyloid-β pathology. The direction of effect was not uniform across primary studies at any stage, and the number of independent samples contributing to each comparison ranged from none to six. Methodological heterogeneity limited direct cross-study comparability. Conclusions: Resting-state EEG microstate parameters show graded group-level differences across clinically defined SCD, MCI, and AD dementia samples. Because the evidence base is predominantly cross-sectional and only four of the 26 studies confirmed underlying AD pathology biologically, these findings should be interpreted as preliminary group-level neurophysiological correlates rather than as established diagnostic biomarkers or as evidence of early-detection or clinical-staging utility. Longitudinal cohorts and diagnostic-accuracy studies in AT(N)-stratified samples are required before any clinical application can be considered. Full article