Search Results (36,436)

Background/Objectives: This study proposes CMT-BUSNet, a hybrid architecture integrating CNN, Mamba, and Transformer branches for breast ultrasound tumor segmentation with built-in explainability. Methods: CMT-BUSNet employs a CNN-anchored hierarchical parallel encoder where Mamba and Transformer branches process CNN-derived features in parallel, fused through an Adaptive Feature Fusion Module (AFFM) with Dense Nested Decoder and Boundary-Aware Composite Loss. Five-fold cross-validation on BUS-BRA (N = 1875) compared nine architectures under identical protocols, plus nnU-Net v2 trained with its default self-configuring protocol as a benchmark. External evaluation used the BUSI dataset (N = 647). Results: CMT-BUSNet achieved DSC = 0.9037 ± 0.0047 on BUS-BRA with higher boundary delineation metrics than nnU-Net v2, which was trained under a different self-configuring protocol (B-IoU: 0.611 vs. 0.557; HD95: 10.07 vs. 13.54 pixels), despite nnU-Net’s marginally higher DSC (0.9108). On BUSI, CMT-BUSNet (DSC = 0.6709) yielded higher scores than nnU-Net (0.5579) across all metrics under zero-shot transfer, though the two methods were trained under different protocols. Training-based ablation confirmed each component’s contribution, and quantitative XAI validation demonstrated attribution faithfulness (nEAR = 2.82×) and uncertainty–error correlation (r = 0.39). Conclusions: CMT-BUSNet achieves competitive accuracy with higher boundary metrics, preliminary cross-dataset transferability, and built-in interpretability relative to nnU-Net (noting different training protocols). Internal validation folds are image-disjoint but not guaranteed to be patient-disjoint, which should be considered when interpreting the reported metrics. Multicenter validation is required before clinical deployment. Full article

The rapid expansion of seaweed aquaculture has profound impacts on coastal ecosystems, yet the lack of long-term, high-precision spatiotemporal monitoring methods has constrained systematic understanding of aquaculture dynamics and their environmental effects. This study integrated Landsat (1984–2025) and Sentinel-2 (2015–2025) imagery with an attention-enhanced U-Net deep learning model to achieve 41 years of continuous monitoring of seaweed aquaculture in the Dongtou Archipelago, Zhejiang Province, China. The model achieved high extraction accuracy for both Landsat and Sentinel-2 aquaculture areas (F1 scores of 0.972 and 0.979, respectively). On this basis, the cultivation zones were further classified into Porphyra sp. and Sargassum fusiforme cultivation areas by incorporating local aquaculture planning and field survey data. Results showed that the aquaculture area underwent three developmental stages: slow initiation (1984–2000, <3 km²), rapid expansion (2001–2015, 3–8 km²), and high-level fluctuation (post-2015, typically 8–20 km²), reaching a peak of ~30 km² during 2018–2019. Long-term retrieval of water quality parameters revealed that the decline in total suspended matter (from ~80 to 60 mg/L) and chlorophyll (from ~3 to 2 μg/L) within aquaculture zones was significantly greater than that in non-aquaculture areas, providing direct observational evidence for local water quality improvement by appropriately scaled aquaculture. Meanwhile, sea surface temperature showed a sustained increasing trend, with extremely high-temperature days (≥25 °C) exhibiting strong interannual variability, posing potential thermal stress risks to cold-preferring seaweed species. The NDVI (Normalized Difference Vegetation Index) and FAI (Floating Algae Index) indices effectively captured aquaculture phenology (seeding, growth, maturation, harvest), with their interannual peaks exhibiting an inverted U-shaped correlation with corresponding yields (R = 0.82 and 0.79, respectively, based on quadratic regression fitting), preliminarily demonstrating the potential of remote sensing in indicating density-dependent effects. This study systematically demonstrates the comprehensive capability of multi-source satellite remote sensing in long-term dynamic monitoring, environmental effect assessment, and yield relationship analysis of seaweed aquaculture, providing key technical support and scientific basis for aquaculture carrying capacity management and ecological risk prevention in island waters. Full article

A suicide epidemic exists among young U.S. veterans, with risk especially elevated in the first year of transition for the 200,000 servicemembers exiting the military annually. The VA Veteran Sponsorship Initiative (VSI) is a public–private-partnership between federal and community partners that aims to decrease suicides by providing a VA-certified volunteer peer sponsor and connection to community services. Onward Ops is a key community-based national program that enrolls, matches and manages the relationship between servicemembers and sponsors. A prior randomized controlled trial showed that the effectiveness of community interventions can be enhanced when augmented by an Onward Ops sponsor. In preparation for national implementation, we conducted a quasi-experimental, matched-cohort pilot to evaluate the feasibility of an adapted VSI protocol and then assessed effectiveness. The adaptations were executed using the Framework for Reporting Adaptations and Modifications-Enhanced between April 2021 and April 2023. The formative results supported the feasibility of the adaptations to enable proactive enrollment on military installations and expand data infrastructure, partnerships, peer sponsors, and VA clinical services. We then assessed the effectiveness for outcomes not studied in the original VSI trial for active-duty soldiers who enrolled between April and December 2023. After nearest-neighbor matching, the sample included 551 VSI participants and 551 soldiers transitioning as usual. The point-probability contrast or risk differences from the conditional logistic regression model indicated that the VSI caused a statistically significant increase in VA primary care utilization of 0.198 and a statistically significant decrease in suicide attempts of −0.019, both assessed 10 months post-military discharge. The study demonstrated the utility of public–private-partnerships, peer-sponsorship programs and enhanced VA services to support servicemembers during transition. Full article

This study is centered on REE distribution in several minerals exhibiting exceptionally rare mineralogical and chemical compositions in the 1.8 Ga albite-enriched granite (AEG) in Madeira. This is a peralkaline A-type granite and corresponds to the Madeira Sn-Nb-Ta-cryolite (REE, Th, U, Zr, Li) world-class deposit (195 Mt) (Amazonas, Brazil). The REE mineralization ranks among the major deposits associated with alkaline and peralkaline magmatism in intracontinental and extensional anorogenic environments in terms of tonnage and grades. However, with respect to REE paragenesis and structure, it differs from all other known REE deposits. The REE mineralization (xenotime, gagarinite, fluocerite, thorite, pyrochlore, zircon, fluorite, and cryolite) is disseminated and zoned. In addition, in the central part of the deposit, there is a massive hydrothermal cryolite body, whose feasibility for REE extracting has been demonstrated. The evolution of rare earth minerals followed a precise order, with minimal formation of compound minerals and minerals with compositions distinct from their typical occurrences. Small pegmatites very rich in xenotime and gagarinite occur in the core AEG. These characteristics are due to the very high F activity in the magma, buffered by cryolite crystallization, to progressive, undisturbed crystallization from the margins toward the center, and to minimal CO₂ activity. The alteration of primary REE minerals by F-rich hydrothermal fluids, the origin of these fluids, and the formation of secondary REE minerals are also discussed. Full article

Lung nodule detection remains a critical challenge in clinical diagnostics due to the small size, weak contrast, and high background interference of nodules in CT scans. To address these issues, a novel deep neural network architecture, termed Daisy-Net, is proposed. This model incorporates dual attention mechanisms and inter-scale feature perception, consisting of two primary components: the Parallelized Patch and Spatial Context Aware (PPSCA) module and the Omni-domain Multistage Fusion (OMF) module. The PPSCA module enhances the extraction of fine-grained textures and boundary information through multi-branch patch perception and spatial attention. The OMF module employs omni-domain feature fusion and progressive stage-wise supervision to improve robustness and discrimination under complex conditions. The lung nodule detection task is formulated as a two-dimensional segmentation problem and evaluated on the LUNA16 dataset. In the post-binarization comparative evaluation, Daisy-Net achieves the best overall performance among all compared methods, with an Intersection over Union (IoU) of 81.41, a Dice coefficient of 89.75, a precision of 95.34, a sensitivity of 84.78, and a specificity of 99.9974. These findings indicate the model’s strong capability in detecting small pulmonary nodules accurately and reliably. Full article

Adverse childhood experiences (ACEs) are a public health concern in the United States. Using the 2019 National Survey of Children’s Health, this cross-sectional secondary analysis examined associations between cumulative ACEs (0, 1, 2, and 3 or more) and three health behaviors among children ages 6 to 17, including physical activity, recreational screen time, and sleep. Interaction models were also estimated by child sex and race/ethnicity (White non-Hispanic, Black non-Hispanic, and Hispanic) to assess whether these associations differed across groups. Nearly half of children experienced at least one ACE, and about one in eight experienced three or more. In adjusted models, higher numbers of ACEs were associated with a lower likelihood of meeting recreational screen time guidelines and sleep recommendations, while no statistically significant association was observed for meeting physical activity recommendations. Interaction analyses by child sex and race/ethnicity found no statistically significant differences in these associations across groups. These findings suggest that children with higher numbers of ACEs may be less likely to meet recommended sleep and recreational screen time guidelines, underscoring the potential value of trauma-informed strategies that strengthen sleep routines and healthy media practices. Full article

Early detection of dental caries in children is critical to prevent irreversible tooth damage and guarantee optimal oral health outcomes. However, interpreting pediatric panoramic radiographs throughout the mixed dentition stage remains a very challenging task due to overlap in anatomical structures and developmental variability. This complexity underscores the need for well curated, representative datasets that enable the development of reliable computer-aided diagnostic models. Herein, this study introduces the Mixed Dentition Orthopantomogram Dataset, a newly developed, publicly available dataset of children that was carefully labeled by dental specialists to identify proximal and occlusal caries regions in the range of 3–12 years. To evaluate the dataset’s applicability for artificial intelligence research, we benchmarked it using both classification and segmentation models. A patch-based classifier achieved an average AUC of 0.89 and Recall 0.85 in distinguishing healthy and carious regions. For segmentation, we evaluated U-Net and Attention U-Net with multiple loss functions, and the Attention U-Net trained with Focal loss achieved the best Dice score of 0.94. Collectively, these findings support the dataset’s utility for pediatric caries analysis and demonstrate the viability of deep learning approaches for mixed dentition panoramic imaging. Full article

Synthetic Aperture Radar (SAR) imagery is inherently degraded by multiplicative speckle noise, rendering traditional superpixel methods—which rely on hard assignment and uniform initialization—suboptimal for boundary preservation. This study proposes a complexity-aware superpixel generation framework featuring differentiable soft-assignment optimization. The approach employs an F-LGRP (Fusion of Local Gradient Pattern Representation) feature descriptor that fuses regional gradient statistics via Gaussian filtering to suppress speckle, coupled with a complexity-driven recursive quadtree initialization strategy yielding non-uniform seed density. A U-Net architecture predicts soft pixel–superpixel association maps within a 9-neighborhood constraint, supervised by a multi-objective loss integrating edge information reconstruction and boundary feature reconstruction. Comprehensive evaluations on simulated and real SAR images (WHU-OPT-SAR and Munich) demonstrate that the proposed method achieves state-of-the-art performance across Boundary Recall, Undersegmentation Error, Compactness, and Achievable Segmentation Accuracy compared to SLIC, SNIC, Mean-Shift, PILS, and SSN. Validation on downstream segmentation tasks further confirms superior accuracy and computational efficiency, establishing the framework as an effective solution for end-to-end SAR image analysis. Full article

To optimize the stiffness matrix structure for frequency-domain elastic wave forward modeling in 2D VTI (transversely isotropic with a vertical symmetry axis) media—thereby reducing memory consumption and improving computational efficiency—we simplify the conventional 25-point finite-difference scheme to derive a 17-point frequency-domain finite-difference scheme. This approach reformulates the finite-difference operators for the partial derivatives and acceleration terms in the elastic wave equations, reducing the number of grid points involved in the computation by $30\%$ compared to the 25-point scheme. The optimized matrix construction leverages sparse matrix storage techniques, decreasing memory usage by approximately $27\%$. Numerical validation, conducted using a double-layer VTI medium model and the Marmousi model with three major faults and an anticline containing limestone layers at the base of the faults, demonstrates that the 17-point finite-difference scheme maintains comparable accuracy while requiring $14\%$ less computation time and featuring a $25\%$ reduction in nonzero elements within the impedance matrix. Comparisons of wavefield snapshots and receiver components (horizontal component U and vertical component V) support this conclusion. These improvements enable the use of more efficient iterative solvers. Full article

The ongoing substance use crisis in the United States involves a broad range of illicit and prescription drugs, including opioids, stimulants, sedatives, and various psychoactive and non-psychoactive compounds. Traditional surveillance methods rely on self-reported data, which could lead to bias and recall inconsistencies. Wastewater-based epidemiology has emerged as a powerful, non-invasive tool for monitoring community-level drug use, offering near real-time estimates and the potential to serve as an early warning system. However, challenges such as analyte degradation, wastewater variability, and matrix effects can affect data quality and comparability across regions. This study presents a standardized, practical workflow for multi-drug (n = 52) detection in wastewater, aiming to minimize analyte loss and improve reproducibility. Composite samples were collected from multiple U.S. cities, transported on ice, and extracted using solid-phase extraction. Extraction efficiencies were compared using Oasis Hydrophilic-Lipophilic-Balanced and Mixed-mode Cation-Exchange (MCX) cartridges, with the MCX sorbent providing complementary reversed-phase and cation-exchange interactions that enabled the retention of chemically diverse compounds across multiple drug classes. Analysis was performed with an Ultra-High-Performance Liquid Chromatography system coupled to a triple quadrupole mass spectrometer, in which the instrument parameters and critical methodological considerations, including sample handling, transport, column selection, and method validation, are detailed. This work contributes to the development of a robust, scalable protocol for multi-drug surveillance in wastewater, supporting timely, data-driven public health responses and informing national drug policy efforts. Full article

Background/Objectives: Wound infections cause significant morbidity, yet current diagnostics rely on time-consuming microbial culture. Volatile organic compounds (VOCs) from bacterial metabolism offer potential for early diagnosis. This study aimed to validate the volatile metabolites profiled by gas chromatography–ion mobility spectrometry (GC-IMS) combined with machine learning for rapid identification of wound infections and certain bacterial infections. Methods: Headspace of clinical wound samples were analyzed using GC-IMS. Volatile metabolite profiles were compared between infected and non-infected groups and between Escherichia coli (E. coli)-positive and negative samples. Partial least squares discriminant analysis (PLS-DA) and Mann–Whitney U test were used for preliminary screening with variable importance in projection (VIP) > 1 and p-value < 0.05. Three machine learning algorithms, namely support vector machine (SVM), logistic regression (LR), and random forest (RF), were trained on the selected features for classification, using 5-fold cross-validation with 10 repeated runs. Model performance was assessed using key evaluation metrics, including accuracy, sensitivity, specificity, the area under the curve (AUC) and feature importance ranking to identify the most relevant biomarkers. Results: A total of 19 volatile metabolites associated with clinical wound samples were identified. The RF model achieved 90.15% sensitivity and 0.91 AUC for bacterial infection detection. For E. coli identification, LR reached 85.35% sensitivity and 0.89 AUC. Potential volatile metabolic biomarkers including elevated 3-methyl-1-butanol, 2-methyl-1-butanol, and ethyl hexanoate for identifying bacterial infection were selected through the cross-validation results of the three algorithms. Conclusions: Untargeted metabolomics by GC-IMS effectively captures infection-specific volatile metabolic signatures in complex wound samples. Integration with machine learning enables rapid, high-accuracy diagnosis of bacterial infections and E. coli identification at point of care. This approach addresses clinical metabolomics translational challenges by providing a portable and cost-effective method, potentially reducing antibiotic misuse through more timely and targeted therapy. Full article

Metal artifacts generated by dental implants significantly degrade cone-beam computed tomography (CBCT) volumes, obscuring critical anatomical structures and compromising diagnostic precision. To address this, an unsupervised deep learning framework has been proposed for Metal Artifact Reduction (MAR) utilizing a Cycle-Consistent Adversarial Network (CycleGAN) optimized for high-fidelity restoration. Unlike supervised methods that rely on unattainable voxel-aligned paired datasets, the proposed approach leverages an unpaired dataset of approximately 4000 images, curated from the public ToothFairy dataset. The architecture integrates U-Net-based generators and PatchGAN discriminators, specifically tuned to mitigate generative hallucinations and preserve morphological integrity. Quantitative benchmarking on a held-out test set demonstrates a 34.6% improvement in the Blind/Referenceless Image Spatial Quality Evaluator (BRISQUE) score, a substantial reduction in Fréchet Inception Distance (FID) from 207.03 to 157.04, and a superior Structural Similarity Index Measure (SSIM) of 0.9105. The framework achieves real-time efficiency with a 3.03 ms inference time per slice, effectively suppressing artifacts while preserving anatomical detail. Expert validation confirms high fidelity; however, to ensure reliability in extreme cases, the architecture is recommended as a clinical decision-support tool under human-in-the-loop oversight. By enhancing diagnostic clarity via a scalable software pipeline, this study provides a robust solution for high-fidelity dental implant imaging. Full article

Investigation of deposystems, sediment routing, and basin architecture during Gondwana breakup refines understanding of Permian–Cretaceous landscape evolution in the central Andes. New chronostratigraphic and provenance constraints from the Eastern Cordillera and Subandean Zone of Bolivia (19–22°S) are based on U-Pb geochronology of detrital and volcanic zircons and ⁴⁰Ar/³⁹Ar dating of interbedded basalts. A discontinuous <2 km-thick Permian–Cretaceous succession records deposition in fluvial, lacustrine, alluvial fan, eolian, and shallow marine environments. Stratigraphic correlations indicate alternations between isolated half-graben subbasins and regional, non-compartmentalized basins. Detrital zircon age spectra from 18 sandstones document sediment recycling from western orogenic and magmatic arc sources and eastern cratonic basement. Synextensional successions of Early Triassic, Early Jurassic, and mid-Cretaceous age were sourced mainly from the west, including Carboniferous and Devonian rocks, while post-extensional fluvial and eolian systems were derived chiefly from the eastern craton. Variations in thickness, facies, and mafic magmatism reflect alternating extensional and neutral tectonic regimes, with localized synextensional subsidence potentially linked to extensional collapse, mantle plume activity, and South Atlantic opening. Comparison with Andean regions in Peru and Argentina indicates that episodic extension and post-extensional thermal subsidence accompanied subduction along the western margin of South America during Gondwana-Pangea breakup. Full article

The rheological behavior of rock masses governs long-term stability, yet the time-dependent properties of grouted structural planes remain insufficiently quantified. Graded shear creep tests were conducted on artificially split sandstone structural planes with controlled grout thicknesses, complemented by scanning electron microscopy (SEM), to clarify creep evolution and long-term shear strength. The results show that the total shear creep displacement of grouted specimens exhibits limited sensitivity to grout thickness, while the ratio of long-term to theoretical shear strength increases by approximately 10% at a grout thickness of 2 mm; this strengthening effect, however, diminishes at greater thicknesses. Moreover, the creep rate evolution of grouted specimens differs fundamentally from that of ungrouted specimens, with about 60% of grouted samples exhibiting an accelerated creep stage characterized by a U-shaped rate curve. The failure mode shifts from asperity-controlled slip in ungrouted structural planes to damage concentrated at the grout–rock interface in grouted specimens. SEM observations further reveal that micro-defects at this interface initiate and propagate cracks, ultimately governing the macroscopic creep failure process. Overall, this study establishes an isochronous curve-based method for determining long-term strength and demonstrates that interface micromechanics critically control the long-term performance of grouted rock masses. These findings provide practical guidance for grouting reinforcement in underground engineering. Full article

This study investigated the effects of computer-assisted writing on the written language production of secondary school students with Specific Learning Difficulties (SLD), particularly dyslexia. Writing is a complex cognitive process requiring the coordination of spelling, lexical retrieval, syntactic organization, transcription, and revision, areas in which students with SLD often experience persistent difficulties. The study compared handwritten and computer-based texts produced by 40 students with SLD and 20 students without learning difficulties using a counterbalanced design, with an interval of approximately two weeks between the two writing sessions. In the handwriting condition, students used printed reference materials, whereas in the computer-based condition they had access to general-purpose digital tools, including spell-checkers, electronic dictionaries, online resources, and word-processing software. Written texts were evaluated using the Spelling Accuracy Index and holistic scores assigned by independent raters. Data were analyzed using descriptive statistics and non-parametric tests (Mann–Whitney U and Wilcoxon signed-rank tests). The findings revealed statistically significant improvements in favor of computer-based writing for both groups, with particularly strong gains among students with SLD. Computer-written texts demonstrated higher spelling accuracy and received higher evaluation scores, indicating improved performance in the assessed writing outcomes. The findings suggest that computer-assisted writing may support written language production in secondary school students with SLD, particularly in relation to spelling accuracy and overall text evaluation, and may offer a useful avenue for more inclusive writing instruction. Full article