Search Results (10,492)

Human activity recognition (HAR) lies at the core of digital healthcare applications that monitor different types of physical activity. Traditional HAR methods often struggle to adapt to variable-length, real-world activity data and to generalise across cohorts (e.g., from young to old cohorts). Thus, the aim of this study was to investigate HAR using wearable sensor data, with a particular focus on cross-cohort evaluation. Each dataset included two accelerometers (right thigh and lower back) sampling at 50 Hz, capturing a range of daily-life activities that were annotated using video recordings from chest-mounted cameras synchronised with the accelerometers. Neural networks were trained on young cohorts’ data and tested on old cohorts’ data. The effects of network architecture, sampling frequency and sensor location on classification performance were investigated. Network performance was evaluated using accuracy, recall, precision, F1-score and confusion matrices. The gated recurrent unit architecture achieved the best performance when trained solely on young cohorts’ data, with weighted F1-score of 0.95 ± 0.05 and 0.93 ± 0.05 for young and old cohorts, respectively, resulting in a highly generalizable method. Classification performance across multiple sampling frequencies was comparable. The thigh-mounted sensor consistently achieved higher performance than the lower back sensor across activities except lying. Furthermore, combining datasets significantly improved performance on the old cohort (weighted F1-score: 0.97 ± 0.02) due to increased variability in the training data. This study highlights the importance of network architecture and dataset composition in HAR and demonstrates the potential of neural networks for robust, real-world activity recognition across age-defined cohorts, specifically between young and old cohorts. Full article

This study evaluates whether a multimodal retrieval-augmented generation (RAG) pipeline originally developed for early modern woodcuts can be effectively transferred to the domain of medieval manuscript illumination. Using a dataset of Wenzelsbibel miniatures annotated with Iconclass, the pipeline combined page-level image input, LLM description generation, vector retrieval, and hierarchical reasoning. Although overall scores were lower than in the earlier woodcut study, the best-performing configuration still substantially surpassed both image-similarity and keyword-based search, confirming the advantages of structured multimodal retrieval for medieval material. Truncation analysis further revealed that many errors occurred only at the deepest Iconclass levels: removing levels raised precision to 0.64 and 0.73, with average remaining depths of 5.49 and 4.49 levels, respectively. These results indicate that the model’s broader hierarchical placement is often correct even when fine-grained specificity breaks down. Taken together, the findings demonstrate that a woodcut-oriented RAG pipeline can be meaningfully adapted to manuscript illumination and that its strengths lie in contextual reasoning and structured classification. Future improvements should incorporate available textual metadata, explore graph-based retrieval, and refine Iconclass-driven pathways. Full article

Lunar dark mantle deposits (DMDs), formed by explosive volcanic activity on the Moon, are typically composed of glass- and iron-rich pyroclastic materials, with slight variations in color, crystallinity, and TiO₂ concentration by region. This paper proposes a method for identifying DMDs using the YOLOv8 deep learning model, enhanced by the introduction of a multi-scale feature extraction (MSFE) module with an attention mechanism, which improves the model’s ability to detect targets at different scales. First, a DMD dataset was constructed using Lunar Reconnaissance Orbiter (LRO) data, with manual annotations of DMD regions and lunar image slicing to optimize computational efficiency. The YOLOv8 architecture, with the incorporated MSFE module, was then used to improve model accuracy in complex terrain. The experimental results showed that the improved DM-YOLO model achieved a precision (P) of 83.9%, a recall (R) of 83.2%, and a mean average precision (mAP@0.5) of 84.2%, representing increases of 15.2%, 14.4%, and 14.0%, respectively, over those obtained with the original YOLOv8 model. The predicted results were preliminarily verified using FeO abundance data and further confirmed by analysis of M³ spectral absorption features, showing strong consistency with known DMDs in terms of both chemical composition and mineralogical characteristics. Observations showed that DMDs were located primarily in the low- and mid-latitude regions of the Moon, with most deposits found in the lunar highlands. The findings suggest that the DM-YOLO model has significant potential for providing technical support for lunar exploration and resource development, particularly for identifying small-scale features that are difficult to annotate. Full article

Semantic segmentation in remote sensing images, which is the task of classifying each pixel of the image in a specific category, is widely used in areas such as disaster management, environmental monitoring, precision agriculture, and many others. However, traditional semantic segmentation methods face a major challenge: they require large amounts of annotated data to train effectively. To tackle this challenge, few-shot semantic segmentation has been introduced, where the models can learn and adapt quickly to new classes from just a few annotated samples. This paper presents a comprehensive review of recent advances in few-shot semantic segmentation (FSSS) for remote sensing, covering datasets, methods, and emerging research directions. We first outline the fundamental principles of few-shot learning and summarize commonly used remote-sensing benchmarks, emphasizing their scale, geographic diversity, and relevance to episodic evaluation. Next, we categorize FSSS methods into major families (meta-learning, conditioning-based, and foundation-assisted approaches) and analyze how architectural choices, pretraining strategies, and inference protocols influence performance. The discussion highlights empirical trends across datasets, the behavior of different conditioning mechanisms, the impact of self-supervised and multimodal pretraining, and the role of reproducibility and evaluation design. Finally, we identify key challenges and future trends, including benchmark standardization, integration with foundation and multimodal models, efficiency at scale, and uncertainty-aware adaptation. Collectively, they signal a shift toward unified, adaptive models capable of segmenting novel classes across sensors, regions, and temporal domains with minimal supervision. Full article

We have developed a tool that allows us to easily visualize evolutionary variation via complementary multiple sequence alignments and frequency-based stacked Sequence Logos. This tool, TOTEMS (hisTogram of evOluTionarily consErved aMino acidS), visualizes conserved regions in a multiple sequence alignment within regions of a three-dimensional structure that share similar degrees of evolutionary conservation as revealed in ConSurf output data. Unlike Sequence Logos that illustrate the relative frequency of individual amino acid residues (as in MSAViewer), or moving window averages that focus on properties such as hydrophobicity or electrical charge (as in CATH), TOTEMS can help users discriminate degrees of evolutionary conservation in adjacent positions within a three-dimensional structure. Thus, we offer a tool that serves to complement pre-existing visualization applications such as ConSurf, MSAViewer, and CATH. TOTEMS and its source code are freely available. Full article

The inference of unstructured text semantics is a crucial preprocessing task for NLP and AI applications. Word sense disambiguation and entity linking tasks resolve ambiguous terms within unstructured text corpora to senses from a predefined knowledge source. Wikipedia has been one of the most popular sources due to its completeness, high link density, and multi-language support. In the context of chatbot-mediated consumption of information in recent years through implicit disambiguation and semantic representations in LLMs, Wikipedia remains an invaluable source and reference point. This survey covers methodologies for entity linking with Wikipedia, including early systems based on hyperlink statistics and semantic relatedness, methods using graph inference problem formalizations and graph label propagation algorithms, neural and contextual methods based on sense embeddings and transformers, and multimodal, cross-lingual, and cross-domain settings. Moreover, we cover semantic annotation workflows that facilitate the scaled-up use of Wikipedia-centric entity linking. We also provide an overview of the available datasets and evaluation measures. We discuss challenges such as partial coverage, NIL concepts, the level of sense definition, combining WSD and large-scale language models, as well as the complementary use of Wikidata. Full article

Phages, the most abundant entities on Earth, exhibit a complex interplay with bacteria, especially within environmental biofilms, resulting in an ecological arms race. This study investigates the interaction between phages and bacteria in the drains of beef-processing plants using high-throughput sequencing and metagenomic analysis. Metagenomic data collected from 75 drain samples from beef-processing plants were analyzed to investigate phage–bacterial interactions. First, assembled contigs were screened to identify viral sequences, which were then taxonomically annotated to determine the viral composition, including phages. Functional annotation of these viral sequences provided information about the viral genes and their roles in bacterial interactions specifically associated with attack and counterattack of bacteria. In parallel, bacterial contigs were examined to identify genes associated with antiphage defense systems, providing insights into the strategies adapted by bacteria to resist phage infection. Taxonomic annotation of viral sequences from the bulk metagenomic data revealed the presence of phages targeting Pseudomonas, Klebsiella, and Enterococcus. The higher abundance of Pseudomonas phages aligns with our previous study, where Pseudomonas was identified as the dominant bacterial genus, suggesting potential copersistence of phages and their hosts. Functional annotation of phage contigs revealed infective and lysis-related genes, highlighting their potential role in bacterial attack. Conversely, bacterial contigs encoded antiphage defense systems, including CRISPR-Cas, restriction–modification, and other defense-related genes. The study also uncovered the presence of anti-CRISPR proteins in phages, suggesting a counterattack on the bacterial defense. These findings provide evidence for phage attack, bacterial defense, and phage counterattack and may showcase the ongoing coevolutionary arms race between phages and bacteria. While this evidence looks promising, these results remain preliminary and further studies are needed to validate these findings. Still, this study provides a foundational understanding of bacteria–phage coexistence in beef-processing plant drains and paves the way for further explorations of these intricate interactions and their possible applications in controlling pathogenic microorganisms within biofilms. Full article

Real-time object detection on resource-constrained edge devices remains a critical challenge in precision agriculture and autonomous systems, particularly when integrating advanced multi-modal sensors (RGB-D, thermal, hyperspectral). This paper introduces FEGW-YOLO, a lightweight detection framework explicitly designed to bridge the efficiency-accuracy gap for fine-grained visual perception on edge hardware while maintaining compatibility with multiple sensor modalities. The core innovation is a Feature Complexity Descriptor (FCD) metric that enables adaptive, layer-wise compression based on the information-bearing capacity of network features. This compression-guided approach is coupled with (1) Feature Engineering-driven Ghost Convolution (FEG-Conv) for parameter reduction, (2) Efficient Multi-Scale Attention (EMA) for compensating compression-induced information loss, and (3) Wise-IoU loss for improved localization in dense, occluded scenes. The framework follows a principled “Compress, Compensate, and Refine” philosophy that treats compression and compensation as co-designed objectives rather than isolated knobs. Extensive experiments on a custom strawberry dataset (11,752 annotated instances) and cross-crop validation on apples, tomatoes, and grapes demonstrate that FEGW-YOLO achieves 95.1% mAP@0.5 while reducing model parameters by 54.7% and computational cost (GFLOPs) by 53.5% compared to a strong YOLO-Agri baseline. Real-time inference on NVIDIA Jetson Xavier achieves 38 FPS at 12.3 W, enabling 40+ hours of continuous operation on typical agricultural robotic platforms. Multi-modal fusion experiments with RGB-D sensors demonstrate that the lightweight architecture leaves sufficient computational headroom for parallel processing of depth and visual data, a capability essential for practical advanced sensing systems. Field deployment in commercial strawberry greenhouses validates an 87.3% harvesting success rate with a 2.1% fruit damage rate, demonstrating feasibility for autonomous systems. The proposed framework advances the state-of-the-art in efficient agricultural sensing by introducing a principled metric-guided compression strategy, comprehensive multi-modal sensor integration, and empirical validation across diverse crop types and real-world deployment scenarios. This work bridges the gap between laboratory research and practical edge deployment of advanced sensing systems, with direct relevance to autonomous harvesting, precision monitoring, and other resource-constrained agricultural applications. Full article

Sympatric species share identical geographical spaces, climatic conditions and survival pressures. Comparative chloroplast genomes among Anna and Lysionotus sympatric species enable exploration of genome-wide evolutionary dynamics of sympatric species. In this study, we assembled and annotated 10 complete chloroplast genomes, representing sympatric species distributed along the Sino-Vietnamese border. We conducted a comparison of chloroplast genomes, characterized their adaptive evolution and used multiple methods to clarify their phylogenetic relationships. Key findings included the following: 1. The number of CDs, rRNA and tRNA varied among different species, whereas they were relatively conserved between the two genera; 2. psaB-psaA, trnL-UAG and ndhD-psaC were identified as potential molecular markers for Anna species, with clpP and ycf1 proposed as effective molecular markers for Lysionotus species; 3. the types of simple sequence repeats (SSRs) and large sequence repeats (LRSs) showed a higher conservation in Lysionotus compared with Anna; 4. the codon usage preferences of the two genera showed convergent evolutionary trends and natural selection played a dominant role, with ycf1 and atpH being confirmed as significantly positively selected genes; 6. phylogenetic analyses using multiple approaches (ML, BI and NJ) consistently verified that Anna and Lysionotus each formed a well-supported monophyletic group. This study offers molecular insights into adaptation and differentiation patterns among distinct plant genera inhabiting the same extreme habitat. Full article

Background: Nanoparticles are a promise for wound-healing therapies. However, its lack of efficacy/safety represents a real challenge for therapeutic use. Objectives: To overcome these problems, the ethanolic extract of Casimiroa edulis leaves, previously reported for its anti-inflammatory, antibiotic, and antioxidant activities, was characterized by FIA-ESI-FTICR-MS and encapsulated in biodegradable nanoparticles for potential wound-healing therapies. Methods:Casimiroa edulis-loaded nanoparticles (CE-NP) were prepared using the rapid emulsion-diffusion method and characterized by their particle size distribution, molecular interactions, charge, morphology, pH, physical stability, and antioxidant and occlusive effects. Results: A total of 40/34 ions in positive/negative electrospray ionization modes were obtained from the extract exploration analysis and were putatively annotated by accurate mass against databases with an error tolerance ≤10 mDa. The most abundant compounds showed the following order: tetramethylscutellarein > rutin > S-usnate > lactose > eugenol derivative > rotenone. While polyphenols predominated, carbohydrates, depsidones/other phenolics, etc., were also detected. The solid/spherical nanoparticles observed by TEM were obtained with a blend of acetone:methyl ethyl ketone (75:25) as the organic phase, producing a unimodal particle size (169.30 ± 1.30 nm; PdI = 0.08 ± 0.03). The encapsulation/loading percentages were 57 ± 0.74/1.62 ± 0.02%, ensuring an entrapment of half the extract, as observed in the FTIR studies. The light backscatter profiles show minimal differences, indicating physical stability correlated with the Z potential (−9.45 ± 1.73 mV). The antioxidant activity of the extract/nanoparticles at 40 µg/mL was 17.27 ± 2.86/16.73 ± 1.28 μg/mL, two-fold higher than that previously reported for sapote seeds. Conclusions: Biodegradable CE-NP with suitable characteristics were obtained for the first time, representing a preliminary proposal for wound healing. Efficacy studies are required. Full article

This study aimed to explore genetic variants associated with serum albumin (ALB) levels in Chinese winter sports athletes using genome-wide association analysis (GWAS) and to investigate potential regulatory mechanisms using bioinformatics annotation. A total of 382 Chinese winter sports athletes were recruited. ALB levels were compared between elite and non-elite athletes. GWAS was conducted using PLINK v1.9, with ALB as the phenotype and sex, age, and principal components as covariates. Associated SNPs were annotated using GTEx and SNPnexus. No significant differences were observed in ALB levels between elite and non-elite male or female athletes, and ALB levels in all groups followed a normal distribution. We identified 113 SNPs reaching a suggestive significance threshold (p < 1 × 10⁻⁵), with per-variant variance explained estimates (7.11–11.76%) reflecting model fit within this cohort. A stepwise regression model highlighted nine candidate SNPs that together explained 51.1% of ALB variance in the study sample. Functional annotation suggested that several variants show eQTL or sQTL signals in tissues relevant to ALB biology (e.g., liver and kidney), and pathway enrichment analyses implicated amino acid and hormone metabolism. Overall, these findings are hypothesis-generating; independent replication in additional and ancestry-matched cohorts (and follow-up functional studies) is required to confirm the robustness of the associations and clarify causal mechanisms. Full article

Strain 11B20 was isolated from a commercial field of maize (Zea mays L.) located in the Yaqui Valley, Mexico. The draft genome sequence revealed a genomic size of 3,759,824 bp, 41.6% G + C content, 973,288 bp N50, 2 L50, and 29 contigs. According to the 16S rRNA gene, strain 11B20 belongs to the genus Bacillus. Genome annotation revealed 3952 coding DNA sequences (CDSs) grouped into 319 subsystems. Among these, several CDSs were associated with traits related to plant growth promotion, including (i) virulence, disease, and defense (33 CDSs); (ii) iron acquisition and metabolism (28 CDSs); and (iii) secondary metabolism (6 CDSs), among others. In vitro, metabolic analysis (IAA, siderophore biosynthesis; phosphorus solubilization; and tolerance to thermal, hydric, and saline stress) confirmed the genomic background of this strain. Finally, in planta assays showed that the inoculation of Bacillus sp. 11B20 significantly (p ≤ 0.05) increased the root length (48.2%) and root dry weight (35.4%) versus non-inoculated maize plants. Thus, this is the first report of Bacillus sp. 11B20 as a promising beneficial strain for sustainable corn production, and further research is needed to ensure the success of the application of this strain in agriculture. Full article

Gestational diabetes mellitus (GDM) develops silently during early pregnancy, yet its earliest circulating molecular signatures remain poorly defined. In this exploratory biomarker study, we characterized first-trimester circulating microRNA (miRNAs) associated with later GDM using a pool-based small RNA sequencing approach. Using a systematic and unbiased sequencing strategy with locus-level miRNA resolution, we profiled the first-trimester plasma miRNome and prioritized a set of 18 mature miRNAs from among 255 detected species. Set-level functional enrichment analyses based on curated and predicted miRNA–target interactions derived primarily from cellular and tissue-based studies showed annotation-based convergence on pathways related to Ca²⁺ homeostasis, glucagon–insulin regulatory circuits, and PI3K–AKT signaling. Network analysis indicated coordinated associations among these miRNAs and shared target pathways involved in insulin secretion and insulin sensitivity. Key contributors—including miR-29a-3p, miR-29c-3p, miR-146a-5p, let-7a-5p, and miR-182-5p—were linked, through in silico target annotation, to central metabolic regulators such as PTEN, PIK3R1, AKT1, AKT2, and components of Ca²⁺ signaling (ATP2A2, CALM1/3, ITPR1, RYR2). These circulating miRNAs should be interpreted primarily as biomarkers reflecting coordinated metabolic states rather than as direct causal mediators. Most identified miRNAs have not been previously reported in the context of first-trimester GDM, supporting the exploratory and hypothesis-generating nature of this circulating miRNA signature in early gestational metabolic research. Full article

Background/Objectives: Chloroplast genomes are essential for understanding the systematics and adaptive evolution of alpine plants, yet genomic data for high-altitude Delphinium species remain scarce. Delphinium densiflorum, a medicinal plant endemic to the Qinghai-Tibet Plateau, exhibits notable high-altitude adaptations, but its plastome features and evolutionary position are still unclear. This study aims to assemble and characterize its complete chloroplast genome and clarify its phylogenetic placement within Delphinium. Methods: Using Illumina NovaSeq data, we de novo assembled the D. densiflorum plastome, annotated it with CPGAVAS2, and compared it with 12 published Ranunculaceae plastomes. We analyzed IR-boundary dynamics, genome-wide sequence variation, and codon-usage bias and constructed a maximum-likelihood phylogeny based on 69 shared protein-coding genes. Results: The plastome is 154,161 bp (GC 38.24%) with a canonical quadripartite structure, encoding 131 genes (87 CDS, 8 rRNA, 37 tRNA). An IR expansion into the SSC region yields the shortest SSC reported among the compared Delphinium species and produces unique structural variants. Photosynthetic genes are extremely conserved (nucleotide diversity Pi ≤ 0.01), whereas several loci (e.g., ycf1 and psaC) are highly divergent (Pi ≥ 0.05). Codon usage shows a strong bias toward AU-ending triplets. Phylogenetically, D. densiflorum forms a 100%-bootstrap clade with other high-altitude congeners, supporting the non-monophyly of Delphinium. Conclusions: This study delineates the plastome architecture and putative adaptive signatures of D. densiflorum, identifies robust candidate loci for DNA barcoding, and provides molecular evidence for taxonomic revision and conservation strategies in Delphinium. Full article

Bridge maintenance decision-making is challenged by the “data-rich but knowledge-poor” nature of unstructured inspection and maintenance reports. A bridge maintenance knowledge graph (BMKG) construction framework is proposed, developed from a corpus of 275 inspection reports, to enable structured representation of engineering knowledge and decision support. A standards-aligned domain ontology provides semantic constraints for downstream information extraction and organization. Building on this ontology, a RoBERTa–BiGRU–CRF named entity recognition (NER) model is developed, achieving a precision of 90.8%, recall of 93.8%, and a micro-averaged F1-score (micro-F1) of 92.3%. Inter-annotator agreement for the NER annotations was quantified using Cohen’s kappa, yielding κ = 0.86. To avoid the cost of large-scale relation annotation, relations are constructed using interpretable, rule-based constraints. Through manual verification audit of randomly sampled relationship instances under a strict exact-match criterion (i.e., requiring exact matches for entity boundaries, entity types, and relationship types), an overall manual verification rate of 93.67% was obtained. Unlike existing KG methods that rely heavily on annotated data, the BMKG framework integrates ontological constraints with a rule-driven approach, prioritizing interpretability and reducing dependency on large-scale relation labeling. Consequently, the resulting knowledge graph supports semantic retrieval and visual exploration, enabling efficient disease-to-recommendation queries for refined bridge maintenance management. Full article