Search Results (742)

Three-dimensional plant point clouds retain crop architecture, but organ-level phenotyping requires reliable semantic separation of leaves and stems. Plant-GeoAT is a geometry-aware parsing network that encodes local relative-XYZ neighbourhoods before RGB fusion and couples spatial neighbourhoods with feature–space relations for dense point prediction. We evaluated the model separately within the native protocol of a self-built structure-from-motion rapeseed dataset, an image-based soybean dataset, and laser-scanned Pheno4D maize and tomato datasets; these are within-dataset train/test experiments, not cross-dataset transfer or domain-generalisation tests. Across five seeds, mIoU was 92.26 ± 0.28%, 82.50 ± 0.24%, 99.74 ± 0.05%, and 94.75 ± 0.15%, respectively. Maize Stem IoU reached 99.57 ± 0.09%. Adding LLGE increased the four-dataset average mIoU from 66.38% to 85.84%, and the complete LLGE + SSCA model reached 92.31%. On the fixed six-sample test set, exploratory semantic-guided clustering achieved 86.67 ± 7.45% Count Accuracy; structural correctness ranged from 3/6 to 4/6 samples across seeds. Plant-height consistency was assessed independently on all 60 reconstructed rapeseed samples. The results indicate that geometry-to-context encoding produces organ-level semantic units for subsequent phenotyping, while independent instance-labelled datasets and additional growth stages are still needed for trait-level validation. Full article

This paper introduces and employs a Computer-Aided Design CAD-to-CAD policy in the context of developing generative models for Machine Learning (ML)-supported hull-form exploration of design spaces in the context of early phases of the ship-design process, using CAD databases as training datasets. For this purpose, two generative models are constructed, a convolutional Generative Adversarial Network (GAN) and a Variational Autoencoder with a Point Cloud Transformer (VAE-PCT) model, that utilize two distinct training datasets of single-type hulls, one consisting of simple (bulbous-bow-free) ship hulls and another consisting of hulls featuring a typical containership. Both models adopt the same CAD-hull discretization, whose combinatorial structure is equivalent to that of a 3D rectangular matrix, and a normalization that proves beneficial regarding the stability, convergence, and time cost of the training process. The architecture of the VAE-PCT model is characterized by a VAE backbone enriched with stacked Point Transformer blocks in the decoder component. The models are compared in terms of both quality and diversity, using five complementary metrics commonly used in the elevant literature. Finally, the CAD-to-CAD policy is materialized by constructing a pull-back map from the generator’s discrete output to a B-spline surface obtained by lofting a family of 3D curves stemming from fairing the generator’s output. The performance of the pull-back map is tested against the geometric validity of the obtained ship hulls and the statistics of (i) nine non-dimensional geometric coefficients, which are correlated with critical performance and operational Key Performance Indicators (KPIs) for ship design, and (ii) calm-water resistance, using an estimator appropriate for the early ship-design phase. The comparison, involving the training CAD models and 8000 CAD hulls obtained by processing the discrete output of the GAN and VAE-PCT generators, confirm that the statistics of the generated CAD hulls are well aligned with those of the training ones. Full article

Image-based cultivar identification remains challenging in perennial medicinal crops because cultivar-specific datasets are often small and morphological differences can be subtle. This study investigated whether a Siamese network-based hybrid learning framework could support cultivar classification and image-derived phenomic similarity analysis in Panax ginseng under low-data conditions. A total of 347 images representing 22 cultivars across four above-ground image acquisition categories were evaluated using stratified five-fold cross-validation. The framework jointly optimized class-weighted cross-entropy and contrastive losses, and five ImageNet-pretrained backbone architectures were assessed across contrastive margins. ConvNeXt-Tiny with a margin of 0.50 achieved the highest five-fold mean performance, with an accuracy of 64.31 ± 7.23% and a macro-F1 score of 61.66 ± 6.63%. UMAP visualization indicated qualitative reorganization of the embedding space after fine-tuning, while Grad-CAM++ localized model responses mainly to plant structures, including leaf, stem, and fruit regions, rather than broad background areas. Hierarchical clustering of cultivar embeddings further suggested structured phenomic relationships, with cosine distance and average linkage yielding a cophenetic correlation of 0.81 ± 0.08 and Kendall’s τ-b of 0.61 ± 0.06. These findings support the potential of hybrid classification and metric learning as a complementary tool for extracting interpretable phenomic representations from limited ginseng image datasets. The framework may provide supporting image-based evidence alongside conventional morphological cultivar assessment. However, the observed cultivar relationships should be considered exploratory and require validation across environments, developmental stages, independent datasets, and genetic information before broader biological interpretation. Full article

Deepfake videos currently facilitate extensive financial deception, political misinformation, and unauthorized imagery, with anticipated U.S. losses from deepfake-related fraud surpassing $40 billion by 2027; human evaluators accurately recognize high-quality forgeries merely 25% of the time, highlighting the pressing necessity for automated, widely applicable detection mechanisms. Adaptive-depth architectures offer an intriguing alternative to fixed-depth deepfake detectors when the optimal model capacity is indeterminate in advance. This study presents the Adaptive Entropy-Guided ICA State-Space Model Forgery Detector (AEGIS-FD), a deepfake detection framework at the video level that progressively increases its depth from one to eight layers via an entropy-driven growth mechanism, attaining peak validation performance at a depth of six. The design incorporates a three-dimensional spatiotemporal stem, Sinkhorn-normalized manifold-constrained hyper-coupling (mHC) layers for balanced temporal integration, a selected state-space temporal block for sequence depiction, and FastICA-based initialization for newly introduced layers. Evaluated using Celeb-DF v2, AEGIS-FD achieves a test AUC of 0.9600 and a validation AUC of 0.9607, above the performance of a single-layer Mamba SSM baseline (AUC = 0.8355). In comparison to a fixed-depth-6 baseline, the model demonstrates consistent improvements over five random seeds (96.12 ± 0.28 vs. 94.84 ± 0.44 AUC; p = 0.0022), suggesting that adaptive development provides advantages that exceed mere depth. In a zero-shot cross-dataset evaluation—trained on Celeb-DF v2 and assessed without fine-tuning on a FaceForensics++ (FF++) C23 subset comprising 1000 original and 1000 FaceSwap videos—AEGIS-FD achieves an AUC of 89.2 compared to 88.4 for the corresponding baseline (+0.8 AUC), providing initial proof of cross-dataset transferability. These findings suggest that adaptive-depth growth presents a viable approach for detecting deepfakes at the video level, while further validation across various datasets and modification techniques is essential. Full article

Graminoid encroachment in alpine tundra is often reduced to a simple rise in dominant-species cover, leaving open whether this aboveground shift is coupled to belowground soil conditioning. Here, we tracked the encroachment front of Deyeuxia angustifolia (Kom.) Y.L.Chang in the Changbaishan alpine tundra, integrating plant community surveys, growth trait monitoring, litter decomposition, phenology-matched soil physicochemical properties, enzyme activities, and bacterial 16S rRNA sequencing across slightly and largely encroached plots at 2000 and 2200 m. Abundance-weighted plant community composition differed significantly across habitats (p = 0.0003), with D. angustifolia relative cover surging from 21.8–29.9% in S to 89.9–93.0% in L, substantially displacing Rhododendron aureum and reducing community Shannon diversity and Pielou evenness. Repeated-measures mixed models revealed that D. angustifolia exhibited elevation-dependent growth trait variation (elevation × stage × date interactions for height, leaf width, stem diameter, leaf count, and tillers; all p < 0.05): plants at 2000 m produced more tillers with a higher seasonal maximum tiller count (active and seasonal maximum tillers), whereas plants at 2200 m prioritized culm reinforcement and foliar expansion (increased leaf width, length, and stem diameter) alongside reduced leaf and branch numbers. Litter decomposition linked these patterns to belowground change: rates were similar between stages at 2000 m, but at 2200 m largely encroached plots decomposed more slowly (lower k, p < 0.001; higher September mass remaining, p = 0.002) while inorganic nitrogen and enzyme activities rose, indicating active local nutrient transformation despite slower turnover. Bacterial community composition varied mainly with elevation (p = 0.001), with further contributions from encroachment stage and month, reflecting compositional restructuring rather than simple diversity change. Together, these results depict D. angustifolia encroachment as a continuous process, from establishment through local space occupancy to litter decomposition dynamics and associated soil habitat differentiation, that reorganizes both aboveground structure and belowground habitats and offers ecological evidence for forecasting tundra meadowization under continued warming. Full article

Co-based oxides are commonly introduced into porous ceramic membranes to add catalytic activity, but their attachment at the atomic scale remains unclear. We examined a buried Co₃O₄/Al₂O₃ interface formed by sol–gel deposition and thermal conversion. Site-specific focused-ion-beam (FIB) lift-out, scanning transmission electron microscopy with energy-dispersive X-ray spectroscopy (STEM-EDS), and high-resolution transmission electron microscopy (HRTEM) were used to access and characterize the interface. A Co-rich spinel-type domain with a (111) lattice spacing was observed next to Al₂O₃(012). The observations guided density functional theory (DFT) initial models. After structural relaxation, substrate-O-mediated Co-O contacts emerged from both starting geometries: the O-bridged-start model exhibited eight contacts across four Co sites, whereas the non-bridged-start model developed three contacts around one Co site. Around the Co-O-Al linkages, there is a clear manifertation of the interface polarization and charge redistribution, indicated by charge-density-difference and Bader analyses. In both models, projected density of states (PDOS) showed coupling between Co 3d and O 2p states, while integrated crystal orbital Hamilton population (ICOHP) analysis further indicated that O atoms retained Al-O bonds while forming occupied-state Co-O bonds. These results support a representative, laterally distributed Co-O-Al motif as an atomic-scale pathway for chemically attaching the functional oxide to porous alumina. Full article

During cherry-tomato-bunch harvesting, clustered fruits, thin stems, and ambiguous fruit–stem junctions cause missed detections and errors in picking-point localization. To address these challenges, we propose TSP-Net, a structure-aware and geometry-constrained detection network built on YOLOv11n, together with an ROI heat-map regression module for picking-point localization. TSP-Net integrates the Truss-aware Multi-scale Ghost Cross Stage Partial (TMG-CSP) module into the backbone to strengthen structural features of fruit edges, small stems, and clusters. It introduces the Truss-oriented Axial-Local Attention (TALA) module during feature fusion to capture fruit-arrangement direction and local occlusion boundaries. We also design Physics-informed Truss Consistency (PTC) Loss to regularize detections by enforcing consistency with fruit-cluster morphology through constraints on center distribution, inter-fruit spacing, and scale continuity. For picking-point localization, local ROIs are extracted from detection outputs, Gaussian heat maps are predicted, and continuous coordinates are decoded using soft-argmax. In the controlled evaluation, TSP-Net attains precision 81.81%, recall 73.21%, mAP@50 79.53%, and mAP@50:95 64.83%, which exceed the corresponding YOLOv11n results by 0.83, 0.32, 1.29, and 1.03 percentage points, respectively. The model size is reduced from 5.23 MB to 5.08 MB, the parameter count falls from 2.59 M to 2.49 M, and computational cost declines from 6.4 to 6.0 GFLOPs. On the common valid-ROI test set, the proposed heat-map localization yields a mean localization error of 47.2 px, a median error of 24.2 px, and PCK@20, PCK@50, and PCK@100 of 45.1%, 68.4%, and 85.5%, respectively. Under identical ROI conditions, this heat-map method reduces the mean localization error by 11.3 px and raises PCK@20 by 14.9 percentage points compared with direct coordinate regression. In summary, the proposed approach enhances cherry-tomato-bunch detection and 2D picking-point candidate localization while meeting lightweight constraints, offering a structure-aware visual perception solution for greenhouse cherry-tomato harvesting. Full article

Machine learning has become a widely adopted approach for URL-based phishing detection, with many studies reporting F1 scores exceeding 0.95 on benchmark datasets. However, recent adversarial machine learning research has questioned the robustness of these models, suggesting that small input perturbations can severely degrade detection performance. In this study, we argue that a substantial part of this reported vulnerability stems from the way adversarial attacks are evaluated. Specifically, many existing studies assess attacks in the feature space, where feature values are modified directly without ensuring that the resulting samples correspond to valid, functional URLs. To investigate this issue, we conduct a two-stage empirical study using both a benchmark feature dataset and a dataset of real phishing URLs. Crucially, to avoid confounding the attack space with dataset differences, we additionally evaluate both feature-space and problem-space attacks on the same real-URL dataset, using an identical model and manipulable-feature budget. Our experiments reveal a striking contrast between these evaluation settings. While feature-space attacks reduce the detection rate of a Random Forest classifier on the benchmark dataset from 0.96 to 0.36, analogous manipulations performed on real URLs have almost no effect on detection performance, as the most informative signals originate from host-related attributes that are difficult for attackers to manipulate. Building on this observation, we propose a set of robust features that capture stable domain characteristics, including lexical word validity, homoglyph disguises, brand impersonation, subdomain depth, character entropy, and transport-related signals. Incorporating these features substantially improves robustness under adversarial conditions, maintaining phishing detection rates between 0.24 and 0.76 where the lexical-only baseline deteriorates to zero under a non-adaptive attacker, while also increasing the clean-data F1 score from 0.985 to 0.994. We further evaluate an adaptive attacker that explicitly targets the proposed features; although the proposed representation raises the attacker’s cost and helps under moderate attacks, host-derived features remain the only strictly attack-invariant component, so we position the proposed features as a complement to host-based signals rather than a standalone defense. Additional analyses, including model comparison, hyperparameter sensitivity analysis, feature ablation, SHAP-based interpretation, multi-seed confidence intervals, a domain-disjoint evaluation, and host-only evaluation, consistently support the proposed approach. The findings demonstrate that problem-space evaluation provides a more realistic assessment of adversarial robustness than conventional feature-space testing and show that robust feature engineering offers a practical strategy for developing phishing detection systems that remain effective under realistic adversarial conditions. Full article

Math walks, where participants explore physical spaces and make connections to mathematics, have been implemented in a wide variety of ways in different learning settings. Math walks connect place-based learning, the development of interest in STEM fields, and opportunities for students to exercise agency over their STEM learning. However, many enactments of math walks have been highly structured and directive in both the mathematics they promote and in the physical movements students are permitted to engage in as they walk. In the present exploratory, qualitative-dominant, mixed-methods bounded case study, we examine eight secondary students engaging in math walk activities as they participate in a STEM summer camp. We examine the nature and types of mathematical questions they ask during math walks, the ways in which they describe the spatial and epistemic agency they have during the experience, and their attitudes toward mathematics. We found that student-led math walks expose a consequential design problem: how to facilitate spatial and epistemic agency while retaining enough mathematical coherence to support depth and continued learning. The article enumerates implications for researchers, educators, and learning designers. Full article

Early diagnosis is essential for the effective management of cardiovascular diseases (CVDs). Although conventional auscultation is the primary screening method, its reliance on subjective interpretation and susceptibility to clinical background noise have positioned phonocardiogram (PCG) analysis as a key diagnostic tool, making reliable automated interpretation a pressing necessity. Automated pipelines have evolved from handcrafted-feature machine learning to deep learning and transformer-based architectures, but the latter often depend on heavy time-frequency preprocessing and large parameter counts, inflating computational cost and increasing the risk of overfitting on limited or noisy clinical data. We propose MTENet, a Multi-representation Time-series Evidential Network that models a phase-enhanced one-dimensional PCG waveform through a bidirectional Mamba state-space encoder, capturing long-range temporal dependencies with linear-time complexity. Recordings are prepared by a label-independent, record-internal stage that combines an adaptive FFT filter bank with an automatically estimated cardiac-phase gain and returns a waveform of unchanged length and sampling rate, so that the model input remains a time-series rather than a fixed feature representation. The Mamba encoder forms one of three parallel branches, alongside an implicit neural representation (INR) branch for continuous signal modelling and a Mel-spectrogram branch computed on-the-fly within the network for spectral structure. The three streams are merged by a softmax-gated fusion. A multi-scale convolutional stem captures local transient structure, while the bidirectional Mamba encoder models longer-range cardiac rhythm. With approximately 2.58 million parameters, the model captures intricate temporal patterns while distributing representational responsibility across complementary streams. Under record-grouped four-fold validation on the primary HLS-CMDS corpus, MTENet attained an accuracy of 0.9816, a balanced accuracy of 0.9812, and an AUROC of 0.9938 under a leakage-free, record-level nested protocol in which the training epoch is selected on an inner-validation split drawn only from the training partition, so that the outer evaluation fold never informs model selection. Within-dataset evaluation on CirCor DigiScope 2022 yielded an AUROC of 0.9698. Full article

Background: Adolescent boys often lack adequate menstrual education, leaving them susceptible to harmful sociocultural influences and reinforcing discriminatory attitudes. The purpose for this study was to explore adolescent boys’ knowledge, sociocultural beliefs, and attitudes toward menstruation. Methods: A descriptive-phenomenological qualitative study was conducted, involving 25 in-depth interviews with adolescent boys enrolled in senior high school (eligibility age range 14–19 years; achieved sample aged 15–19 years) from five districts in Ghana’s Volta Region. Participants were purposively sampled to ensure diverse perspectives. Semi-structured interviews explored knowledge sources, sociocultural influences, and attitudes toward menstruation. Data were analyzed thematically using Braun and Clarke’s six-phase framework with the aid of MAXQDA software. Results: Adolescent boys’ knowledge of menstruation was fragmented, stemming from family, school, and internet sources. Mothers were frequently cited as key informants, but misconceptions about menstruation persisted. Sociocultural and religious beliefs reinforced menstruation-related taboos, such as restrictions on cooking and entering sacred spaces, shaping discriminatory attitudes. Reactions to menstrual leaks varied from empathy to teasing. Conclusions: The findings highlight the need for inclusive menstrual education targeting boys and their families. Educational programs should include biological, emotional, and social aspects of menstruation to help reduce myths and nurture empathy. Full article

Despite outstanding results, machine learning-based Android malware detection models struggle with concept drift, where rapidly evolving malware characteristics degrade model effectiveness. This study examines the impact of concept drift on Android malware detection, evaluating two datasets and nine machine learning and deep learning algorithms, as well as Large Language Models (LLMs). Various feature types—static, dynamic, hybrid, textual, and image-based—were considered. The results showed that concept drift is widespread and significantly affects model performance. Factors influencing the drift include feature types, data environments, and detection methods. Balancing algorithms help with class imbalance but do not fully address drift, which primarily stems from the dynamic nature of malware. No strong link was found between the type of algorithm used and concept drift; the impact was relatively minor compared to other variables because hyperparameters were not fine-tuned, and the default algorithm configurations were used. The LLM evaluation is treated as an exploratory baseline because the original feature spaces were compressed using PCA to satisfy token-length constraints. Under this setting, LLMs showed promising few-shot performance but remained sensitive to temporal drift. Full article

Chronic musculoskeletal pain remains a leading cause of disability worldwide, driven by progressive degeneration of cartilage, bone, tendon, intervertebral discs, and peripheral nerves. Conventional interventional approaches primarily address symptoms without restoring structural integrity or tissue homeostasis. Regenerative strategies, including platelet-rich plasma, mesenchymal stem cells, and biomaterials, have demonstrated potential but are limited by variability in outcomes, poor cellular survival, and lack of standardization. Exosomes and extracellular vesicles are key mediators of intercellular communication in tissue repair, reproducing many of the paracrine effects of parent cells while offering improved safety and scalability. These nano-sized vesicles regulate inflammation, angiogenesis, extracellular matrix remodeling, and cell survival across musculoskeletal and neural tissues. Importantly, growing evidence suggests that mechanical cues such as compression, shear stress, and tensile loading not only regulate cellular behavior but also shape exosome biogenesis, cargo composition, and functional effects through mechanotransduction pathways involving integrins, ion channels, and YAP/TAZ signaling. This mechanobiology–exosome interface is particularly relevant in interventional pain medicine, where therapeutics are delivered into mechanically active environments such as joints, discs, tendons, and perineural spaces. Mechanical loading conditions may therefore modulate therapeutic efficacy by shaping both endogenous repair processes and the behavior of administered exosomes. The present investigation reviews the current literature and knowledge on mechanotransduction pathways and exosome biology, with a focus on their intersection in musculoskeletal regeneration. We further examine preclinical and clinical evidence supporting exosome-based therapies in osteoarthritis, degenerative disc disease, tendon and ligament injuries, and neuropathic pain states, alongside key translational challenges including heterogeneity in exosome isolation, dosing, biodistribution, and clinical standardization. Finally, we discuss future directions in mechanobiology-guided exosome engineering and their potential integration into interventional pain practice. Full article

Rolling element bearings are critical components in rotating machinery, yet assessing and predicting their reliability under heavy industrial noise remains challenging. Existing methods suffer from three major limitations: (1) single-channel signal processing and single-scale indicators lack robustness against non-stationary noise; (2) classical multi-channel decomposition methods, such as multivariate variational mode decomposition (MVMD), rely on empirical parameter tuning, which frequently leads to over- or under-decomposition; and (3) monolithic deep architectures and homogeneous ensemble models suffer from prediction drift and generalization bottlenecks during long-term temporal extrapolation. To address these issues, this paper introduces an automated framework that combines adaptive multi-channel signal purification with a heterogeneous stacking ensemble (HeteroStack-LR). Unlike conventional MVMD pipelines that fix $[K,\alpha ]$ empirically, the Sequoia Optimization Algorithm (SOA) autonomously determines the globally optimal configuration, achieving a mean SNR of $-2.08\mathrm{dB}$—a $1.88$ to $2.50\mathrm{dB}$ improvement over standard VMD/MVMD baselines—along with up to a $37.1\%$ reduction in computation time. Rather than relying on conventional single-metric intrinsic mode function (IMF) selection, we construct a multi-domain hybrid index integrating the Fault Correlation Factor, energy ratio, and refined composite multiscale dispersion entropy (RCMDE) to robustly identify noise-resistant components, thereby enhancing denoising quality by $22.4\%$ to $32.2\%$ over single-scale criteria. Furthermore, contrasting with linear PCA-based reduction, Diffusive Topology Neighbor Embedding (D-TNE) effectively preserves the nonlinear manifold structure of degradation trajectories in a low-dimensional space. Finally, a heterogeneous stacked ensemble featuring an out-of-fold (OOF) leakage-prevention strategy and a logistic regression meta-learner is designed to suppress prediction drift while avoiding the over-parameterization typical of deep architectures. Experimental results across four bearing datasets demonstrate that HeteroStack-LR achieves a minimal MAE of $0.063$ with a variance of ≤±0.002, outperforming state-of-the-art deep architectures (such as TCN, CNN-LSTM, BiLSTM-Attention, and Transformer) as well as classical baselines (Bi-LSTM, CNN, and LSSVM). Ablation studies confirm that removing SOA and MVMD degrades MAE by $12.7\%$ and $19.0\%$, respectively, validating that the framework’s strength stems not from any isolated module, but from the end-to-end synergistic integration of signal purification and reliability prediction. Full article

This study presents a comprehensive experimental and theoretical investigation into dual-gate field-emission devices fabricated using a standard 0.35 µm CMOS-MEMS process. Two emitter configurations, the concave-tip and triangular-tip, are characterized, and their performance is rigorously analyzed using three-dimensional simulations based on Fowler–Nordheim emission theory. To account for discrepancies between initial designs and fabricated devices, the influence of critical geometric parameters, including tip apex radius, cathode-anode spacing, and tip sharpness, is systematically evaluated regarding emission current and threshold voltage. Compared to the floating-gate baseline (~38 V), dual-gate (DG) operation lowers the threshold voltage by ~70% (~10 V), enhances low-voltage emission over tenfold, and provides a 3.4-fold boost in differential output conductance. Simulation analysis indicates this improvement stems from enhanced electrostatic field distribution governed by the gates. Furthermore, the top gate, due to its proximity to the emitter tip relative to the bottom gate, provides superior control over emission current at lower operating voltages. Three-dimensional simulations corroborate these findings, revealing that minimizing both the tip radius and cathode-anode spacing substantially enhances tunneling electron flow. Additionally, gate voltage sweeps confirm that electron trajectories are effectively directed by electrostatic steering. These findings establish critical design guidelines for integrating field-emission devices into standard CMOS platforms, facilitating the development of on-chip electrostatically controlled electron sources for integrated vacuum microelectronics. Full article