Search Results (79,429)

High-rise LiDAR scanning produces vertically sparse point clouds where upper-layer defects are hardest to detect due to inverse-square ranging law (1/r²) density gradients, noise contamination, and complex geometries. This paper presents PC-TowerNet, a physics-aware AI pipeline that achieves state-of-the-art reconstruction through sequential modules: (1) 50D geometric feature classification outperforming CloudCompare SOR (100% accuracy vs. 91.3% retention); (2) Physics-Constrained VAE (PC-VAE) recovering 28.7 ± 2.1% upper density vs. 8.3 ± 1.7% standard VAE; (3) multi-modal PointNet++/GNN/Transformer fusion; and (4) Bayesian uncertainty maps (ECE = 0.042 ± 0.008). Synthetic tower evaluation (10 × 5 seeds) demonstrates 48.9% surface smoothness improvement and 38.2% volume error reduction over tuned RANSAC baselines, with clear paths to real-data validation. Full article

Building damage and collapse are the main sources of serious casualties and financial losses during earthquakes, which are among the most destructive natural disasters that endanger human life and property. Therefore, quick and precise post-earthquake building damage assessment is essential for risk assessment and emergency action. Convolutional neural networks (CNNs) primarily concentrate on local features and frequently ignore global contextual information within and across buildings, despite the fact that deep learning-based techniques allow automated damage identification. Transformer-based approaches, on the other hand, are good at capturing global dependencies, but their large memory and processing costs restrict their usefulness. As a result, existing networks still struggle to achieve an effective balance between accuracy and efficiency. To address this issue, this study proposes a lightweight and efficient network for post-earthquake building damage assessment. Specifically, we develop a two-stage method based on EfficientViM with an encoder–decoder architecture. In the encoder, Mamba is introduced to extract multi-scale change features with long-range dependencies, leveraging the state space model to preserve global modeling capability while significantly reducing computational complexity. In the decoder, two lightweight modules are designed to further enhance discriminative capability and computational efficiency. The network finally outputs building localization and pixel-level building damage, respectively. Experiments were conducted on four earthquake events from the BRIGHT dataset using a three-for-training and one-for-testing cross-event rotation evaluation strategy. The results demonstrate that LEViM-Net requires only 30.94 M parameters and 27.10 G FLOPs. In addition, for the Türkiye earthquake event, the proposed method achieves an F1 score of 80.49%, an overall accuracy (OA) of 88.17%, and a mean intersection over union (mIoU) of 49.73%. The proposed model enables efficient remote-sensing-based mapping of macroscopic and image-visible building damage, providing timely support for early-stage emergency response. Full article

This study proposes a standards-based Reference AI Business Model Canvas (Reference AI-BMC) that translates the use-case descriptors of ISO/IEC TR 24030 into the nine blocks of the Business Model Canvas, addressing the lack of a structured translation layer between AI standards and business-model design. Using ten selected fields of the ISO/IEC TR 24030 use-case template, a two-round Delphi process derives consensus-based mapping rules from expert judgments; Latent Dirichlet Allocation is used as a field-level semantic analysis to provide interpretive context for the Delphi-derived mappings. Primary mappings are reported as default translation references that met the 80% strict-consensus threshold, secondary mappings as context-dependent relations, and the adjudicated dual-mapping exception A5 (Threats/Challenges → Cost Structure) as a separately documented case. After converting the finalized primary mapping rules into a coding manual, three independent coders applied them to 81 AI use cases; the Layer 1 coding yielded Krippendorff’s α = 1.000, descriptively indicating no observed coder disagreement under the specified coding conditions. The Reference AI-BMC contributes a standards-based, consensus-derived translation layer for systematically organizing AI use cases in business-model terms, offering a structured starting point for early use-case workshops, preliminary portfolio screening, and standards-aware AI service design discussions. Together, these results position the Reference AI-BMC as a standards-based, consensus-derived reference layer for organizing AI use cases in BMC terms, with its applicability bounded by the ISO/IEC TR 24030 descriptor structure and the specified mapping procedure. Full article

Tumor cell heterogeneity and multicellular interactions critically influence drug resistance, recurrence, and prognosis. Here, CPcellsubpopulation, a computational framework integrating scRNA-seq, bulk RNA-seq, and clinical data was developed to identify cancer progression-associated cell subpopulations. Then, the integrated analyses of scRNA-seq and spatial transcriptomics were performed to predict potential interactions, identify critical transcription factors, and predict candidate anticancer drugs. Across nine cancers, we detected cancer progression-associated cell subpopulations significantly linked to prognosis, with consistent patterns across cancer types. In renal cell carcinoma (RCC), we identified conserved metabolic^high UBE2C+ cancer cells linked to poor outcomes, metabolic reprogramming and low differentiation, and PLK1+ NK cells, plasma cells, and CDC20+ macrophages associated with advanced stages and unfavorable prognosis. Spatial mapping revealed spatial association of RCC progression-associated cancer and immune cell subpopulations, suggesting the potential role of the VEGF, GDF, PTN and IL16 pathways in the remodeling of the tumor microenvironment. Gene regulatory network analysis highlighted RAD21 as a key regulator linking metabolism and therapy resistance. This study provides a systematic pipeline to delineate cancer progression-associated cell subpopulations, uncovers metabolic^high UBE2C+ cancer cells as progression-associated tumor cell population, and nominates critical regulators and compounds as therapeutic targets. Full article

With the development of autonomous driving and intelligent connected vehicle technologies, the vehicle cabin is shifting from a simple transportation space to an intelligent mobile space integrating infotainment, interaction, and rest, and passenger comfort has gradually become an important factor affecting user experience, system trust, and perceived safety. Focusing on three categories of cabin environmental factors, namely the acoustic, optical, and thermal environments, this study develops an experimental design and comprehensive modeling method for passenger comfort evaluation. First, controlled single-factor experiments were conducted to establish quantitative mapping relationships between physical environmental parameters and subjective comfort ratings. The analytic hierarchy process (AHP) was then used to determine the weights of each indicator, and a penalty-based aggregation mechanism was introduced to construct a comprehensive comfort evaluation model. Finally, external validation was performed on an independent vehicle platform to examine the model’s applicability and consistency. The results show that acoustic comfort decreases as the sound pressure level increases, whereas optical and thermal comfort exhibit nonlinear behavior with optimal intervals. AHP weight results show that the thermal environment has the highest weight (0.4280), followed by the acoustic environment (0.3305) and the optical environment (0.2415). The external validation results indicate that the proposed model exhibits good predictive consistency across three steady-state operating conditions, with a mean absolute error of 0.122, a root-mean-square error of 0.150, and a Pearson correlation coefficient of 0.960. The findings show that the penalty-based aggregation model can effectively characterize the limiting-factor effect under the joint action of multiple environmental factors, providing a computable and interpretable evaluation framework for intelligent cockpit environmental control and automotive engineering experimental teaching. The conclusions of this study are mainly applicable to the current experimental platform and steady-state operating conditions, and further validation is still required with more vehicle models, dynamic road scenarios, and complex multi-environment factor disturbances. Full article

Background: Multi-objective optimization (MOO) is widely used in engineering design and energy systems to represent trade-offs through Pareto fronts. Yet practical deployment requires moving from a non-dominated set to an implementable preferred design, and this decision step is often treated implicitly. Many studies equate decision support with improved Pareto front generation or visualization, while decision-maker preferences are assumed, weakly specified, or not elicited from stakeholders. Methods: A two-phase scoping evidence synthesis with PRISMA-informed reporting was adopted to map the literature and synthesize explicit Pareto-front decision-support mechanisms. Phase 1 produced a broad evidence map of how Pareto-front decision support is framed and clustered studies by primary contribution, while Phase 2 conducted a focused synthesis of explicit Pareto-front decision-support methods using refined searches in Scopus and SpringerLink. Results: Phase 1 mapped 46 studies; only 10 reported an explicit reproducible Pareto front solution-selection mechanism. Phase 2 included 17 studies and identified four method families: post hoc scoring and ranking, compromise aggregation, interactive preference-guided exploration, and preference elicitation and learning. Conclusions: The literature remains dominated by Pareto front generation and exploration rather than reproducible final solution selection; future work should strengthen preference elicitation, transparency, sensitivity analysis, and uncertainty-aware recommendation stability. Full article

This paper presents a robust and efficient mmWave radar-based human activity recognition (HAR) framework optimized for practical real-time indoor deployment. Addressing computational inefficiencies and limited recognition scopes in existing systems, the framework introduces two core contributions: Multi-class Spatio-Temporal Network (MuST-Net), a lightweight, multi-class network, and an online detection process for enhanced temporal stability. MuST-Net utilizes a hybrid 2D convolutional neural network and temporal convolutional network architecture to recognize seven distinct classes, significantly broadening the system’s recognition repertoire. The online detection process implements a novel sliding-window post-processing chain that employs an activity-buffering mechanism, which maintains temporal continuity and effectively suppresses spurious detections at activity boundaries. Experimental results demonstrate the superior performance of our unified framework, attaining over 98.6% accuracy for multi-class classification by MuST-Net and achieving at least 97% accuracy for activity detection and a crucial 100% recall for fall detection. Robustness is validated across three distinct indoor environments and nine subjects—with two of the three sites entirely unseen during training—confirming strong generalization under installation, environment, and subject variations. Full article

Open-vocabulary semantic segmentation (OVSS) aims to achieve pixel-level object segmentation guided by arbitrary natural language descriptions. Although pre-trained vision–language models (VLMs) have significantly advanced the development of OVSS, their reliance on the Vision Transformer (ViT) architecture imposes a fundamental constraint on dense prediction. Specifically, the absence of hierarchical downsampling in ViT-based VLM results in single-scale representations that trade spatial localization for global semantics. To address these issues, this paper proposes a hierarchical boundary-constrained decoding network for OVSS, called CLIP-HBD. Our approach leverages VLM semantic priors to reconstruct multi-scale features and introduces a boundary-constrained decoding strategy to refine edge details. Specifically, CLIP-HBD leverages a ConvNeXt-based backbone alongside a hierarchical adaptation mechanism to fuse multi-layer VLM features, generating a comprehensive multi-scale representation. To address the issue of boundary inaccuracy, we perform explicit boundary prediction based on multi-scale representations, where the resulting boundary maps are subsequently transformed into structural constraints to steer the decoder’s focus toward boundary regions. By integrating structural constraints with hierarchical features, the decoding process effectively maintains semantic consistency and restores precise object boundaries. Extensive experiments demonstrate that CLIP-HBD achieves superior performance in both segmentation precision and boundary quality across multiple benchmarks. Full article

Background: Antigenic drift limits the protective efficacy of influenza vaccine. Glycosylation of hemagglutinin (HA) represents a promising immunofocusing strategy that enhances neutralizing antibody responses by masking immunodominant non-neutralizing epitopes. Methods: B-cell epitopes of influenza viruses were retrieved from the Immune Epitope Database and were mapped onto the HA structure of A/Puerto Rico/8/1934 (H1N1). Structure-guided analysis identified residues 136 and 137 as candidate sites for N-linked glycosylation (NLG). Single-site mutants (136NLG and 137NLG) were generated using reverse genetics and evaluated for stability, receptor binding, viral replication, and immunogenicity in a murine model with inactivated whole-virus vaccines. Results: Both mutants exhibited increased thermostability at 42 °C. Glycosylation reduced the HA–sialic acid affinity, resulting in decreased viral adsorption and internalization efficiency in MDCK cells, and delayed viral replication at low multiplicity of infection (MOI). In vivo, all vaccine groups provided complete protection against lethal challenge; notably, the 136NLG group exhibited reduced weight loss, indicating improved protective efficacy compared with wild-type (WT). Conclusions: Targeted glycosylation at residue 136 in the HA head domain effectively enhances the viral stability and elicits a 1.78-fold increase in hemagglutination inhibition titer (GMT) relative to the WT, thereby improving vaccine performance. These findings establish a rational and structure-based design strategy for developing more stable and effective influenza vaccines. Full article

Lactoferrin is a naturally occurring bioactive glycoprotein that is part of the body’s innate immune system and has essential roles in iron metabolism, microbial defense, inflammation regulation, and tissue repair. It supports the natural regulation of iron bioavailability in skin and hair follicles, helping to reduce excess free-iron-driven oxidative stress while preserving levels of necessary iron for cellular functions. Lactoferrin promotes cell regeneration by increasing proliferation across in vitro systems, stimulating wound healing in scratch assays, and boosting matrix production in fibroblast models. Lactoferrin can also modulate inflammatory signaling involved in skin and hair physiology by providing balanced cytokine suppression, suggesting potential value in cosmetic or dermatological applications. Here, we present the first focused summary of lactoferrin’s role specifically in skin and hair biology, distinguished from prior reviews in systemic or multi-system broad health contexts. We link mechanistic insights with clinical and preclinical evidence and uniquely map molecular functions to dermatologic and trichologic outcomes. We also provide an overview of clinical skin studies that have explored lactoferrin as a supportive agent in conditions such as acne, and highlight that, despite mechanistic plausibility, there are no existing available reports of well-controlled human clinical trials leveraging lactoferrin for hair-focused outcomes. In summary, we propose lactoferrin as not just an anti-inflammatory molecule, but also as a microenvironment stabilizer, and particularly relevant for hair and skin support as an alternative to pharmacological interventions. By addressing both established and underexplored applications, this review provides a translational framework for clinical development and provides a comprehensive rationale behind leveraging lactoferrin for hair and skin epithelial health. Full article

Large-scale hydrogen storage is expected to play a critical role in balancing the variability of renewable energy systems, particularly those driven by wind power. However, the combined influence of storage capacity and deliverability on supply reliability remains insufficiently characterized. This study investigates the reliability limits of hydrogen storage systems operating under variable hydrogen production and time-varying demand. A dimensionless modeling framework is developed to map system performance across a wide range of storage capacities and deliverability levels. The results reveal a clear transition between reliable and unreliable operating regimes. Reliable operation requires a minimum deliverability level approximately equal to the mean hydrogen production rate, corresponding to a value of about 1.05–1.10 times the average production across the range of intermittency conditions considered in this study (from moderate to highly variable production). Below this threshold, increasing storage capacity alone cannot prevent supply shortfalls. Once this threshold is exceeded, further increases in deliverability provide diminishing returns and storage capacity becomes the dominant factor governing reliability. In this regime, the required storage capacity approaches a plateau on the order of 10–30 days of average hydrogen throughput, depending on the level of production variability. The proposed regime-based framework provides a practical tool for evaluating storage feasibility and guiding preliminary capacity design in renewable hydrogen systems. Full article

Background: Children with obesity-related asthma exhibit poorer symptom control and more frequent exacerbations than their normal-weight peers, but the underlying metabolic mechanisms are unclear. This study aimed to identify drivers of obesity-related asthma through untargeted plasma metabolomic and lipidomic profiling. Methods: Plasma was obtained from normal weight (NW) asthmatic (n = 95) and non-asthmatic (n = 67) and overweight/obese (OO) asthmatic (n = 99) and non-asthmatic (n = 100) children (6–17 years). We assessed metabolic and lipidomic differences between asthmatics and controls within each BMI group using orthogonal partial least squares discriminant analysis (OPLS-DA), examined overlap with the adult Qatar Biobank cohort, and mapped metabolic–clinical interactions using Gaussian Graphical Models. Results: In the fitted OPLS-DA models, separation between asthmatic and control groups was stronger in the NW group (R²Y = 0.72/0.52) than in OO (R²Y = 0.65/0.63) children. Asthma was associated with altered tricarboxylic acid (TCA) intermediates, ether-linked phosphatidylethanolamines, and sphingomyelins (SM) in NW, and with phosphatidylcholines, lysophosphatidylcholines, and phosphatidylethanolamines in OO. Integrating metabolomic, lipidomic, and clinical data revealed connections between altered SMs and interleukins, and TCA intermediates and electrolytes, all associated with elevated leptin in NW. An increased residual volume to total lung capacity ratio in OO was associated with phospholipid shifts. The overall dynamics in lipid metabolism with asthma, conditioned on BMI, was also observed in the adult Qatar Biobank cohort. Conclusions: Among NW children with asthma, we found enhanced TCA cycle activity and inflammation linked to altered SM metabolism, whereas in OO, the findings suggest oxidative stress arising from chronic obesity-related inflammation. These data reveal BMI-specific metabolic mechanisms of pediatric asthma that might inform precision approaches to disease management. Full article

To mitigate the structural amplification of random false alarms during multi-pulse repetition frequency (Multi-PRF) ambiguity resolution, this paper proposes a general waveform synthesis optimization framework based on Bezout’s Identity and Genetic Algorithm (Bezout-GA). By leveraging Bezout’s Theorem, the framework establishes an analytical mapping between the Greatest Common Divisor (GCD) topology of transmission parameters and system-level false alarm boundaries. It is mathematically demonstrated that the uncontrolled inflation of the Least Common Multiple (LCM) in traditional coprime-based strategies leads to severe “spatial over-issuance” of false alarms, a phenomenon particularly exacerbated in heavy-tailed K-distributed sea clutter. The proposed two-stage hybrid paradigm employs a genetic algorithm for global multi-objective search, followed by local number-theoretic refinement via the Extended Euclidean Algorithm to strictly satisfy hardware constraints. Simulations across X-band and L-band scenarios confirm the framework’s superior spectral generalizability. Results indicate that the Bezout-GA optimized waveform achieves a 4.1-fold reduction in expected false alarm volume at the cost of a negligible 0.1% clear-region sacrifice. Notably, in extreme K-distributed clutter ($\nu =0.1$), the framework reclaims an equivalent signal-to-clutter-and-noise ratio (SCNR) gain of up to 3 dB in the L-band, significantly outperforming traditional coprime and maximum clear-region benchmarks. Overall, this study provides a number-theoretic perspective for analyzing spatial false alarm mechanisms and serves as a methodological reference for future investigations into robust Multi-PRF waveform optimization. Full article

Restless legs syndrome (RLS) is a common sensorimotor disorder with limited treatment options and incompletely understood pathophysiology. Genome-wide association studies have identified numerous risk loci, but translating these findings into causal genes and therapeutic targets remains challenging. We performed a proteome-wide association study (PWAS) integrating RLS genome-wide association study (GWAS) data from FinnGen with two brain pQTL datasets (ROSMAP and Banner). We validated the identified proteins using TWAS, SMR, and colocalization analyses using brain pQTL and eQTL datasets. To further investigate peripheral protein associations, we performed SMR using plasma pQTL data from the UK Biobank Pharma Proteomics Project (UKB-PPP). We also conducted a phenome-wide association study (PheWAS) to screen for potential off-target effects of the prioritized genes, followed by drug prediction using DSigDB and molecular docking. PWAS identified GLO1, along with GRWD1 and MAP2K5, as significantly associated with RLS. GLO1 was identified by brain-based SMR (p = 0.0001), colocalization (PP.H4 = 0.96), TWAS (p = 0.048), and was confirmed by plasma-based SMR (p = 3.16 × 10⁻⁹) as the only protein associated with RLS. PheWAS analysis, without associations for 783 non-RLS phenotypes, confirmed the specificity of GLO1. Among 27 predicted GLO1-targeting compounds, Gambierol had the strongest binding affinity (−8.3 kcal/mol). This proteogenomic study identifies GLO1 as a prioritized causal gene and promising drug target for RLS, combining brain and plasma data to provide new insights into pathogenesis and candidate drug development. Full article

Vision–Language–Action (VLA) policies promise flexible long-horizon manipulation, but deployment under domain shift requires both reliable uncertainty estimates and a workable runtime-assurance policy. We study a model-agnostic uncertainty-calibrated safety-gating wrapper that estimates online failure risk and routes control among policy execution, pause-and-reobserve, and a fallback planner. Using a cleaned and consistently aggregated benchmark pipeline, we evaluate two long-horizon manipulation tasks in NVIDIA Isaac Sim 5.0 under lighting, texture, occlusion, sensor, and combined shifts. Relative to an ungated VLA baseline, calibrated gating improves mean shifted success from 57.5% to 77.2% and reduces aggregate expected calibration error from 0.303 to 0.100. The largest success gains occur under occlusion and combined shift, including improvements from 48.3% to 85.2% on the drawer task and from 59.4% to 87.8% on clutter sort. The results also expose a systems trade-off: an aggressive uncalibrated threshold baseline attains stronger raw success and collision metrics, but requires nearly twice as many interventions per shifted episode (21.6 vs. 11.5). The main contribution is, therefore, an empirical characterization of the reliability–intervention trade-off created by calibrated supervision, not a claim that the calibrated supervisor is universally the best terminal controller. We frame calibrated gating as a better-calibrated, lower-intervention supervisor that materially improves robustness relative to an ungated VLA while revealing the open problem of mapping calibrated risk into efficient intervention policies. Additional threshold-sensitivity, signal-diagnostic, overhead, and residual-failure analyses show that the selected operating point is meaningful but not universal: the calibrated risk threshold captures most shifted failures in retrospective logs, yet residual contacts still arise during pause and fallback states. These findings provide controlled simulation evidence for trustworthy VLA supervision under distribution shift and clarify the reliability–intervention frontier that future embodied-control systems must navigate. Full article