Search Results (3,988)

Using home sensors to provide agitation warnings for community-dwelling dementia patients without ongoing clinical supervision can enable their carers to intervene early during the agitation latent period, thereby reducing unnecessary hospital admissions and harmful events. Most existing studies are based on behavioral, sleep, and physiological data from the previous 24 h to predict patients’ agitation events, which fail to fully capture patients’ recent behavioral details. In this study, monitoring data from the previous 8 × 24 h for dementia patients are used to achieve in-depth mining of patients’ living habits. Furthermore, to address the common clinical problem of extreme imbalance between agitation and normal samples (agitation samples often are extremely scarce), we designed a two-stage agitation early warning model based on symmetric autoencoders and a spatio-temporal network, dubbed SAS-Net. In the first stage, we randomly sample 80% of normal samples and employ multiple symmetric autoencoders to perform feature transformation and pseudo-label construction, and then pre-train the spatio-temporal learning network composed of convolutional networks and a gated Multilayer Perceptron. It aims to learn the patient data’s intrinsic structure, underlying patterns, and effective feature extraction methods. In the second stage, we freeze the parameters of the spatio-temporal learning network and use a balanced dataset consisting of the remaining 20% of normal samples and all agitation samples to reconstruct and fine-tune the top fully connected classifier to improve the recognition performance of agitation samples. The two-stage strategy resolves the problem of ineffective training faced by deep learning models on imbalanced datasets. The experimental results demonstrate the effectiveness of the proposed SAS-Net for agitated behavior early warning. Full article

Gas pipeline leakages in urban underground utility tunnels pose a severe threat to public safety. Leakages of varying aperture sizes trigger differentiated risks of diffusion and explosion; thus, achieving precise identification of leakage hole size has become a critical issue in safety management. To address the difficulty of traditional methods in effectively separating the acoustic features of different leakage levels within complex utility tunnel environments, this paper proposes a gas pipeline leakage risk level identification method based on a lightweight Spatial–Temporal Graph Neural Network (ST-GNN). First, relying on a real utility tunnel simulation platform, acoustic signals under different pressures and leakage hole size are collected, and time-frequency magnitude features are constructed through Short-Time Fourier Transform (STFT). Furthermore, each acoustic sample is independently converted into a graph with STFT time frames as nodes, where temporal neighborhood edges and K-nearest neighbor edges jointly encode local dynamics and non-local spectral similarities. This transforms unstructured acoustic signals into graph-structured data that embodies spatial–temporal coupling relationships. Building upon this, a lightweight Chebyshev graph convolutional network is designed to progressively extract discriminative features strongly correlated with leakage levels using multi-layer convolution. Experimental results on the actual utility tunnel simulation platform dataset demonstrate that the proposed method achieves excellent performance in a three-level leakage classification task. The t-SNE visualization reveals the effective separation of features, progressing from complete mixing in the input layer to distinct separation in the output layer. Through multiple training statistics and ablation experiments, the impact of dataset size and the number of network layers on the identification performance is analyzed, validating the robustness of the proposed model under limited samples and the effectiveness of its lightweight structure. This provides a feasible solution for the automated and refined identification of gas pipeline leakage levels in underground utility tunnels. Full article

Point cloud segmentation constitutes a core task in 3D computer vision. However, prevailing models suffer from inherent limitations, including the absence of polarity correlation (i.e., spatial attribute-containing features derived from the separation and calculation of positive/negative correlations within point cloud query–key pairs), inefficient feature fusion, loss of fine-grained geometric details, and excessive computational complexity in self-attention mechanisms. These deficiencies constrain both the performance and practical deployment of such models. To address these challenges, the Polarity-Aware Attention and Feature Dynamic Fusion Network (PA-DFNet) is proposed in this paper. Built upon the PointNet++ framework, PA-DFNet replaces the original Multilayer Perceptron (MLP) with a Polarity-Aware Network (PAN). The PAN enhances key semantic interactions by explicitly separating positive and negative correlations from point cloud query–key pairs, generates adaptive neighborhood weights via integration with a linear attention mechanism, and introduces a learnable power function to perform nonlinear scaling of attention, thereby improving the model’s structural perception capability. Additionally, a Point Cloud Feature Dynamic Fusion (PFF) module is proposed to enable adaptive fusion of encoder–decoder features, preserving rich geometric details. Experimental results demonstrate that, on the ModelNet40 classification task, the overall accuracy (OA) and mean accuracy (mAcc) of PA-DFNet are improved by 2.4% and 2.2%, respectively, compared with PointNet++. On the S3DIS semantic segmentation task, PA-DFNet achieves an mAcc of 72.8% and a mean Intersection over Union (mIoU) of 66.2%, while exhibiting a shorter training time than Point Transformer. In summary, PA-DFNet achieves an optimal balance between segmentation performance and efficiency by effectively controlling the number of model parameters and computational complexity. Full article

Therapeutic proteins have emerged as a cornerstone of modern medicine due to their high specificity and strong biological effects. However, delivering these proteins poses significant challenges due to their instability, susceptibility to enzymatic breakdown, low permeability, and reliance on invasive parenteral routes. Buccal drug delivery is a promising non-invasive alternative, offering quick systemic absorption while avoiding gastrointestinal degradation and hepatic first-pass metabolism. Three-dimensional (3D) printing as a fabrication method has further enhanced the potential of buccal delivery, enabling precise dosage control, multilayer structures, and patient-specific customization. This review focuses on the current state of the traditional and 3D-printed buccal film platforms using different printing methods for protein delivery, and critically analyzes protein stability challenges, and formulation strategies. The discussion further highlights emerging proof-of-concept studies. Full article

Breakthrough innovation—central to industrial competitiveness and the ongoing clean-energy transition—remains persistently constrained by information homogenization and weak cross-domain integration in single-layer innovation networks. Technology Innovation Composite Networks (TICNs) have therefore been advocated as dual-layer platforms coupling knowledge and collaboration networks, yet the cross-layer mechanism through which they generate breakthrough outputs has not been specified. This paper specifies and tests how knowledge-layer structural holes open access to heterogeneous information that must cross into the collaboration layer to be recombined into breakthroughs. Two distinct boundaries shape the outcome. Inventors’ finite cognitive processing capacity makes integration returns decay along an inverted U-shape; separately, excessive degree and closeness centrality drive the collaboration layer into homogenization and localization, narrowing the range of structural holes it can productively absorb and shifting the breakthrough peak toward lower structural-hole levels. Together, they delineate an optimal cross-layer integration zone. Using panel data on 10,681 patents, 948 inventors, and 5631 inventor-year observations from new energy (2004–2018), a fixed-effects negative binomial model confirms the inverted U-shape and the steepening, peak-shifting moderations of degree and closeness centrality; a Lind–Mehlum test places the turning point inside the observed data range, and negative binomial (robust SE), Poisson and zero-inflated Poisson specifications—together with a stricter top-1% breakthrough threshold—yield consistent results. The study moves multilayer network research from structural description toward mechanism-level identification and offers actionable network-design guidance. Full article

The present work explores the modelling and design of Interface Acoustic Wave (IAW)-based delay lines in SiO₂/ZnO (4 µm)/SU-8/SiO₂ multilayer stacks and demonstrates that, by properly tailoring the acoustic wavelength and the SU-8 layer thickness, IAW delay lines can achieve performances comparable to, and in some cases superior to, those of conventional Surface Acoustic Wave (SAW) delay lines based on SiO₂/ZnO (4 µm) structures. In particular, the proposed devices exhibited untuned insertion losses down to 12 dB, propagation losses as low as 0.052 dB/λ, and electromechanical coupling coefficients K² approaching 4%, exceeding those calculated for the corresponding SAW devices. The obtained results support the feasibility of compact, high-performance, and potentially packageless acoustic-wave devices for future telecommunications and sensing applications, especially in harsh or contamination-prone environments. Full article

Gastrointestinal (GI) cancers account for a quarter of all cancer cases worldwide and are responsible for a third of cancer deaths. One of the characteristic features of GI tissue is its multilayered structure, which, in addition to multiple scattering, complicates optical spectral analysis. The use of spectroscopic diagnostics and photodynamic therapy for the detection and treatment of GI cancer is a rapidly developing field. The method proposed in this paper for layer-by-layer optical properties assessment, suitable for real-time clinical application to the walls of hollow organs, allows us to calculate the absorbed dose layer by layer. This paper proposes a method for recording spectral data in two geometries, diffuse reflectance and transmission, using light delivery from both the external and internal surfaces of the gastrointestinal tract wall. Layer-by-layer assessment of optical properties was performed using a developed algorithm based on the inverse adding–doubling method with initial optical properties values determined using the modified two-stream Kubelka–Munk model with the accuracy equal to 86 ± 13%. The method was approved in clinical conditions. Based on the results of the work, the developed method for assessing the optical properties of multilayered biological tissues exhibited sufficient speed and accuracy for in vivo application to personalize laser-induced therapy by correction of the laser dose. Full article

Accurately evaluating how aero-engine performance supports upper-level capability remains a challenging issue in the digital planning, demonstration, and design of complex equipment systems-of-systems. Existing studies mainly rely on two-level analyses at the subsystem and system-of-systems levels, which are insufficient to characterize the cross-level transmission relationships among the aero-engine, aircraft performance, and overall capability. To address this limitation, this paper proposes a multilayer network-based contribution evaluation method for aero-engines oriented toward digital equipment planning and demonstration. First, a three-layer evaluation index system is constructed, including the overall capability layer, the aircraft performance layer, and the aero-engine performance layer, based on the OODA loop concept and aviation physical constraints. This provides a structured and traceable basis for cross-level requirement decomposition and scheme evaluation. Second, by integrating expert prior judgment with mechanism-based sensitivity analysis, the interrelationships among indicators at different layers are quantified, and a multilayer evaluation index network is established. Third, topological structure analysis is employed to identify key indicators in the aero-engine layer, and a cascading propagation model is introduced to evaluate the supporting roles and contribution rates of both individual indicators and the overall aero-engine layer with respect to the overall capability layer. Simulation results show that the proposed method can effectively reveal the structural characteristics, propagation paths, and dynamic influence patterns of aero-engine-layer indicators within the multilayer network. The proposed method provides methodological support for digital equipment planning, scheme demonstration, design optimization, and capability-oriented decision-making of aero-engines. Full article

Rainfall-index-based disaster insurance is an efficient approach to mitigating hydrometeorological losses. However, conventional premium pricing models generally assume memoryless stochastic dynamics that do not fully capture the long-range dependence inherent in rainfall data. This study develops a hydrometeorological disaster insurance model within a fractional Black–Scholes framework to incorporate long-memory effects. The model is formulated using fractional differential equations and solved semi-analytically by integrating the Daftardar–Jafari Method (DJM) with the Kashuri–Fundo (KF) transform, yielding a closed-form solution expressed in terms of the Mittag–Leffler function. The proposed contract is structured as parametric rainfall insurance with a multi-layer payout mechanism based on percentiles corresponding to minor, moderate, and severe housing damage. The results show that variations in the fractional-order parameter significantly affect premium estimation. In particular,  δ = 0.5 recovers the classical model and tends to generate higher premiums than the fractional model with δ = 0.23153, whereas the model with δ = 0.73153 yields lower premiums. These findings indicate that fractional-order parameterization can accommodate diverse risk characteristics and policyholders’ economic capacities, enabling more adaptive, risk-sensitive premium structures. In line with SDG 13 (Climate Action), the proposed framework offers a climate-responsive disaster-mitigation strategy through accessible, actuarially relevant insurance design.  recovers the classical model and tends to generate higher premiums than the fractional model with , whereas the model with  yields lower premiums. These findings indicate that fractional-order parameterization can accommodate diverse risk characteristics and policyholders’ economic capacities, enabling more adaptive, risk-sensitive premium structures. In line with SDG 13 (Climate Action), the proposed framework offers a climate-responsive disaster-mitigation strategy through accessible, actuarially relevant insurance design. Full article

Under the growing threat of urban heat stress, street canyons play a critical role in shaping the pedestrian thermal environment. While street greenery is an effective mitigation strategy, its performance varies substantially with physical characteristics—such as aspect ratio, street width, and sidewalk width—highlighting the need for spatially adaptive design. This study evaluates the effects of sidewalk width, street orientation, and planting structure on thermal conditions in a humid subtropical climate in Daegu Metropolitan City, Republic of Korea. The analysis focuses on open low-aspect-ratio street canyons (H/W = 0.86 for E–W and 0.43 for N–S orientations). Using a validated ENVI-met (Version 5.6.1) model based on field measurements from Daegu, Republic of Korea, 56 street-greening scenarios were simulated by systematically varying sidewalk width, street orientation, planting rows, spacing, and planting structure. Results show that multi-row planting served as the primary structural framework governing thermal performance. Optimal configurations varied with sidewalk width, with two-row planting for 6 m sidewalks and three-row planting for 10 m sidewalks providing the most effective cooling. The greatest cooling (−2.02 °C) was achieved when optimized multi-row configurations were combined with multi-layer planting. Once optimal multi-row configurations were established, the presence of understory vegetation had a greater influence on thermal improvement than its specific composition, allowing flexibility in understory design. Clear spatial asymmetries were identified, with the highest thermal stress occurring on the north-side sidewalk in E–W streets and the west-side sidewalk in N–S streets. Targeted planting in these locations produced greater cooling benefits than uniform strategies. These findings provide a spatially grounded framework for climate-responsive street greenery and offer practical design guidance, highlighting the need for context-specific, optimized multi-row planting strategies adapted to local urban and climatic conditions. Full article

The vibration characteristics of clamped–clamped Alumina–Steel axially functionally graded (AFG) fluid-conveying Timoshenko pipes are investigated using a physics-based generalized integral transform technique (GITT) benchmark and a multi-layer perceptron (MLP) surrogate trained on GITT data. Parametric GITT sweeps over the power-law gradation index k, dimensionless flow velocity u, and aspect ratio $L/D$ quantify how axial material gradation controls the first two natural frequencies (${\omega }_1$, ${\omega }_2$) and the maximum vibration deflection ($y_M$): increasing k reduces ${\omega }_1$ and ${\omega }_2$; on u-sweeps at $L/D=50$, larger k also increases $y_M$ and lowers the critical flow velocity, whereas on $L/D$-sweeps at $u=3.0$, $y_M$ decreases with k. A feedforward MLP surrogate fitted to $N_{\mathrm{s}}=336$ GITT samples via an interior block-wise train–test split and three independent networks with output-specific preprocessing achieves $R^2>0.99$ on held-out data, with maximum relative errors below 9%, and reproduces representative GITT parametric curves in overlay validation. After one-time offline training, MLP inference is orders of magnitude faster than online GITT runs, enabling large-scale global sensitivity analysis based on Sobol indices, SHAP values, and partial dependence plots; these identify u as the dominant influence on the modal responses, while SHAP ranks k first for ${\omega }_2$. The physics-based GITT and MLP surrogate workflow combines high-fidelity material–structure benchmarking with efficient metamodeling for design optimization, reliability assessment, and sensitivity-driven screening of Alumina–Steel AFG fluid-conveying pipes. Full article

Under lead–bismuth eutectic (LBE) corrosion conditions, the multilayer oxide layer that forms on T91 steel adversely affects on its thermal conductivity. This study systematically conducted corrosion experiments under varying temperatures, durations, oxygen concentrations, and bismuth (Bi) content. By combining microstructural characterization with laser flash measurements of thermal conductivity, the evolution of T91 thermal conductivity under different corrosion conditions was revealed. Based on these findings, molecular dynamics simulations based on the neuroevolution potential (NEP) framework were employed to construct a T91/Fe-Cr spinel/Fe₃O₄ multilayer heterojunction model, enabling precise determination of the intrinsic thermal resistances at the two interfaces. By coupling the interfacial thermal resistances with experimental data, the macroscopic effective thermal conductivities of Fe-Cr spinel and Fe₃O₄ in real corrosion environments were calculated to be 1.68 W/(m·K) and 2.19 W/(m·K), respectively. These values are significantly lower than those reported for pure phases, thus revealing the inhibitory effect of defects and pores in actual oxide layers on heat transport. This research establishes a multiscale analytical method spanning from atomic-scale interfacial thermal resistance to macroscopic heat transfer properties of oxide layers, thereby providing a theoretical basis and data support for the thermal performance evaluation and service life prediction of LFR structural materials. Full article

Radiation monitoring with SRAM-based FPGAs traditionally relies on offset-histogram analysis, which requires a chip-specific calibration campaign at an accelerator before multiple-cell upsets (MCUs) can be discriminated from coincident single-cell upsets (SCUs). The cost and complexity of such calibration restrict the approach to dedicated, beam-test-funded programs. We propose an AI-based on-board classifier that achieves MCU/SCU discrimination directly, without any chip-specific calibration. A lightweight Multi-Layer Perceptron (MLP), trained entirely on synthetic data covering five representative bit-interleaving layouts, is integrated on an AMD Artix-7 XC7A200T FPGA together with per-detection-element telemetry aggregation. The classifier achieves F1 = 0.92–0.97 on structured BRAM layouts when per-chip calibration data are available (calibrated ceiling) and, without any chip-specific calibration, retains F1 up to 0.81 ± 0.02 (held-out, mean over five seeds) on previously unseen layouts with near-perfect recall. A sensitivity analysis across a $20\times$ range of SEU rates and a $4\times$ range of MCU fractions confirms the robustness of the proposed approach. A feature-ablation study identifies an indispensable feature subset, while a comparative evaluation of four alternative classifier architectures (decision tree, support vector machine (SVM), two MLP variants) establishes the reference MLP as the optimal choice. Post-implementation results on the Artix-7 200T show that the MLP-enhanced and calibrated-histogram designs occupy nearly identical FPGA footprints, reframing the choice between them as an operational decision driven by calibration availability rather than by hardware cost. Full article

This paper investigates the effects of disturbance observers on a long-travel precision positioning platform. We propose a multi-layer control architecture, including a disturbance observer, a feedforward compensator, gain-scheduling, and control switching. The platform consists of motor and piezoelectric transducer (PZT) stages to enable nanometre-level accuracy within 10 cm travel ranges. We identified the dynamic models of the stages through experiments and applied them to develop control designs. The PZT stage was equipped with feedforward compensators, a disturbance observer and real-time switching control schemes to achieve robust and precise tracking. On the other hand, we applied gain-scheduling and feedforward compensation to the motor stages to track large displacements. The control effects of the integrated platform were validated through simulations and experiments and demonstrated significant improvements in accuracy and robustness. Finally, the platform was incorporated with two-photon polymerisation to fabricate micro-lenses. This work evaluates the lenses’ optical properties to highlight the advantages provided by the multiple control structure for improving precision and microfabrication applications. Full article

Traditional approaches to urban real estate green value assessment rely heavily on single structured data sources. Such methods often provide limited interpretability and fail to capture multidimensional green attributes accurately. To address these limitations, this study constructs a multimodal assessment framework that integrates image, text, and spatial information. A housing price prediction model is developed based on a Multi-Layer Perceptron architecture. Results show that the proposed method is superior to traditional models (such as the Hedonic pricing model, Ridge regression, and eXtreme Gradient Boosting, as well as single-modality control models). The core evaluation metric, mean squared error, reaches 0.0505 ± 0.0021. SHapley Additive exPlanations analysis shows that the text modality provides the largest contribution to model prediction, accounting for 51.45% of the global contribution. However, this dominance reflects the model’s dependence on textual green signals rather than the establishment of causal relationships. The result may also be influenced by marketing language bias and symbolic sustainability signals. The image modality contributes 38.48%, while the spatial modality contributes 10.07%, indicating a complementary relationship among the three modalities. Green premium analysis confirms that the model achieves higher prediction accuracy for high-priced residences and effectively captures differences in green premium across housing price tiers. This study provides a new technical pathway for real estate green value assessment. Full article