Search Results (623)

The performance of extractive summarization models is often limited by their dependence on human references that may contain inaccuracies or subjective biases. Existing methods typically rely solely on sentence-level supervision, which lacks explicit grounding in the actual semantic content of the source document, thus limiting their robustness. We propose SSF-KW, a novel multi-task learning framework that enhances robustness by jointly optimizing keyword extraction and sentence selection. Our approach is designed to explicitly anchor salience decisions in the document’s intrinsic semantic structure, reducing reliance on potentially noisy labels. To this end, the model employs a shared BERT encoder to represent sentences, and identifies keywords through part-of-speech tagging, semantic similarity analysis, and fine-grained keyword signals with sentence-level representations via a transformer-based fusion module. The entire framework is optimized with a combined loss function that balances both tasks. Comprehensive evaluations on CNN/DailyMail, XSum, and WikiHow demonstrate that SSF-KW consistently outperforms baselines ROUGE-1 scores of 43.27, 25.43, and 30.03, respectively. Ablation studies confirm the contribution of each component, with the word-level module proving especially critical for capturing key concepts in procedural texts like WikiHow. Full article

Music is a near-universal anthropological and sensory phenomenon that engages distributed brain networks and peripheral physiological systems to shape emotion, cognition, sociality, and bodily regulation. Evidence from electrophysiology, neuroimaging, endocrinology, randomized controlled trials, and longitudinal training studies indicates that both receptive and active musical experiences produce experience-dependent neural and systemic adaptations. These include entrainment of neural oscillations, modulation of predictive and reward signaling, autonomic and neuroendocrine changes, and long-term structural connectivity alterations that support affect regulation, cognition, social functioning, motor control, sleep, and resilience to neuropsychiatric illness. This narrative review integrates mechanistic domains with clinical outcomes across major conditions, such as depression, anxiety, schizophrenia, dementia, and selected neurodevelopmental disorders, by mapping acoustic and procedural parameters onto plausible biological pathways. We summarize how tempo, beat regularity, timbre and spectral content, predictability, active versus passive engagement, social context, dose, and timing influence neural entrainment, synaptic and network plasticity, reward and prediction-error dynamics, autonomic balance, and immune/endocrine mediators. For each condition, we synthesize randomized and observational findings and explicitly link observed improvements to mechanistic pathways. We identify methodological limitations, including heterogeneous interventions, small and biased samples, sparse longitudinal imaging and standardized physiological endpoints, and inconsistent acoustic reporting, and translate these into recommendations for translational trials: harmonized acoustic reporting, pre-specified mechanistic endpoints (neuroimaging, autonomic, neuroendocrine, immune markers), adequately powered randomized designs with active controls, and long-term follow-up. Contextual moderators including music education, socioeconomic and cultural factors, sport, sleep, and ritual practices are emphasized as critical determinants of implementation and effectiveness. Full article

Background: Hippocampal neurogenesis in the dentate gyrus persists into adulthood and plays a crucial role in learning and memory. Early-life exposure to low-dose propofol has been reported to enhance neural development in rodent models, but detailed mechanisms remain unclear. To address this gap, we aimed to investigate how low-dose propofol alters neurogenic lineage differentiation, transcriptional programs, and underlying molecular mechanisms within the early postnatal hippocampal neurogenic niche. Results: We conducted an in-depth re-analysis of a published single-nucleus RNA-sequencing (snRNA-seq) dataset from hippocampal tissue of postnatal day 10 (PND10) mice, collected 3 days after low-dose propofol treatment. Uniform Manifold Approximation and Projection (UMAP)-based clustering revealed twelve major cell types, including a population of Ntng1⁺Fxyd7⁺Pcp1⁺ immature pyramidal neurons (imPYR), lacking the mature markers Meis2 and Spock1. Trajectory analysis revealed two neurogenic lineages (granule and pyramidal) and indicated that propofol biases progenitor fate commitment towards the granule lineage. CellChat analysis demonstrated that propofol enhances Neurexin (Nrxn) signaling to neural progenitor cells, suggesting increased synaptic adhesion and maturation. Differential expression analysis (|log₂FC| ≥ 0.26, adjusted p < 0.01) followed by pathway enrichment revealed that propofol upregulates neurogenic maturation pathways—including synaptogenesis, synaptic transmission, dendritic morphogenesis, and memory-related processes—specifically within neural intermediate progenitor cells (nIPC). Conclusions: Together, these findings delineate a coordinated transcriptional and intercellular mechanism by which low-dose propofol reprograms hippocampal neurogenesis during early postnatal development, highlighting progenitor-specific and synapse-oriented processes that may underlie its cognitive-enhancing effects. Full article

Peer assessment has emerged as a crucial approach for scaling evaluation in educational scenarios, fostering learner engagement, critical thinking, and collaborative learning. Nevertheless, traditional aggregation-based and probabilistic methods often fail to capture the intricate relational dependencies among students and submissions, thereby limiting their capacity to ensure reliable and equitable outcomes. Recent advances in graph neural networks (GNNs) offer promising avenues for representing peer-assessment data as graphs. However, most existing approaches treat all relations uniformly, overlooking variations in the reliability of evaluative interactions. To bridge this gap, we accordingly propose ReSAN (Relation-Sensitive Assessment Network), a novel framework that integrates relation-sensitive attention into the message-passing process. ReSAN dynamically evaluates and weights relationships, enabling the model to distinguish informative signals from noisy or biased assessments. Comprehensive experiments on both synthetic and real-world datasets demonstrate that ReSAN consistently surpasses strong baselines in prediction accuracy and robustness. These findings underscore the importance of explicitly modeling evaluator reliability for effectively capturing the dynamics of peer-assessment networks. Overall, this work advances reliable graph-based evaluation methods and provides new insights into leveraging representation learning techniques for educational analytics. Full article

Graph anomaly detection (GAD) aims to identify nodes or edges that deviate from normal patterns. However, the presence of heterophilic edges in graphs leads to feature over-smoothing issues. To overcome this limitation, this paper proposes the multi-view heterogeneity resistant network (MV-GHRN) model, which progressively purifies heterophilic edges through multi-view collaboration. First, to address the noise sensitivity of single predictions, the method computes post-aggregation (PA) scores for both the original graph and its perturbed versions and performs weighted fusion, leveraging the consistency of multiple prediction perspectives to enhance the reliability of heterophilic edge identification. Second, a cosine similarity view is introduced as a complementary structural perspective, with both views independently completing heterophilic edge pruning to clean the graph structure from both topological and feature dimensions. Finally, a cross-view self-distillation mechanism is designed, using the fused predictions from the two purified views as teacher signals to guide the optimization of each view in reverse, correcting feature biases caused by heterophilic edges. Experiments on benchmark datasets such as YelpChi and Amazon demonstrate that the framework significantly outperforms existing methods. For instance, on the YelpChi dataset, MV-GHRN surpasses the best baseline by 16.8% and 5.2% in F1-Macro and AUC, respectively, validating the effectiveness of the progressive multi-view purification mechanism. Full article

Our understanding of Ghrelin, an endogenous ligand of the growth hormone secretagogue receptor 1a (GHSR1a), has expanded from considering it to be a “hunger hormone” to a pleiotropic regulator of whole-body physiology. This review synthesizes the current advances spanning ghrelin biogenesis, signaling, and systems biology. Physiologically, preproghrelin processing and O-acylation by ghrelin O-acyltransferase (GOAT) generate acyl-ghrelin, a high-potency GHSR1a agonist; des-acyl ghrelin predominates in circulation and exerts context-dependent, GHSR1a-independent, or low-potency effects, while truncated “mini-ghrelins” can act as competitive antagonists. The emergence of synthetic ligands, agonists, antagonists, and reverse-agonists has provided the necessary tools to decipher GHSR1a activity. Recent cryo-EM structures of GHSR1a with peptide and small-molecule ligands reveal a bipartite binding pocket and provide a framework for biased signaling, constitutive activity, and receptor partner selectivity. Beyond the regulation of feeding and growth-hormone release, ghrelin modulates glucose homeostasis, gastric secretion and motility, cardiovascular tone, bone remodeling, renal hemodynamics, and innate immunity. Ghrelin broadly dampens pro-inflammatory responses and promotes reparative macrophage phenotypes. In the emerging scholarship on ghrelin’s activity in the central nervous system, ghrelin has been found to influence neuroprotection, stress reactivity, and sleep architecture, and has also been implicated in depression, Alzheimer’s disease, and substance-abuse disorders. Practical and transitional aspects are also highlighted in the literature: approaches for ghrelin stabilization; recent GHSR1a agonists/antagonists and inverse agonists findings; LEAP-2-based strategies; and emerging GOAT inhibitors. Together, structural insights and pathway selectivity position the ghrelin system as a druggable axis for the management of inflammatory diseases, neuropsychiatric and addiction conditions, and for obesity treatment in the post-GLP-1 receptor agonist era. Full article

Establishing robust geochemical baselines in the hyper-arid Atacama Desert remains challenging because of extreme climatic gradients, polymetallic mineralisation, and decades of intensive mining. To disentangle natural lithogeochemical signals from anthropogenic inputs, a region-wide, multi-institutional soil dataset (1404 samples; 32 elements) was compiled. The analytical workflow integrated compositional data analysis (CoDA) with isometric log-ratio transformation (ILR), principal component analysis (PCA), robust principal component analysis (RPCA), and consensus anomaly detection via hierarchical (HC) and spectral clustering (SC), applied both with and without spatial coordinates to capture compositional structure and geographic autocorrelation. Optimal cluster solutions differed among laboratory subsets (k = 2–17), reflecting instrument-specific biases. The dual workflows flagged 76 (geochemical-only) and 83 (geo-spatial) anomalies, of which 33 were jointly identified, yielding high-confidence exclusions. Regional baselines for 13 priority elements were subsequently computed, producing thresholds such as As = 66.9 mg · kg⁻¹, Pb = 53.6 mg · kg⁻¹, and Zn = 166.8 mg · kg⁻¹. Incorporating spatial variables generated more coherent, lithology-aligned clusters without sacrificing sensitivity to geochemical extremes (Jaccard index = 0.26). These findings demonstrate that a reproducible, compositional-aware machine learning workflow can separate overlapping geogenic and anthropogenic signatures in heterogeneous terrains. The resulting baselines provide an operational reference for environmental monitoring in northern Chile and a transferable template for other arid mining locations. Full article

This study aimed to validate a bias estimation framework for low-cost maritime IMUs by applying it to real-world shipborne data. Six estimation methods—including statistical (mean, median), model-based (least squares, cross-correlation), and signal-processing approaches (FFT, Butterworth filter)—were compared. The results demonstrated that the low-frequency Butterworth filter achieved the smallest residuals, with RMS residuals below 0.038 m/s² for accelerometers and 0.0035 deg/s for gyroscopes. In particular, AccX and AccZ residuals converged to 3.04 × 10⁻² m/s² and 2.30 × 10⁻² m/s², respectively, while GyroZ achieved 5.58 × 10⁻⁴ deg/s. Estimated accelerometer biases were 0.0405 m/s² (X-axis) and 0.1615 m/s² (Y-axis), and the optimization successfully converged with an objective function value of 9.314. The findings confirm that the previously proposed bias estimation method, originally validated in simulation, is effective under real-world maritime conditions. However, as ground truth bias values cannot be obtained in shipborne experiments, verification relied on residual statistics and cross-correlation analysis. This limitation has been explicitly stated in the conclusion, and future studies should incorporate sensitivity analyses and controlled experiments to further quantify error sources. Full article

Anxiety and emotional stress are pervasive psychological challenges that profoundly impact human health in today’s fast-paced society. Traditional assessment methods, such as self-reports and clinical interviews, often suffer from subjective biases and lack the capability for objective, real-time evaluation of mental states. However, the integration of physiological signals—including electroencephalography (EEG), heart rate (HR), electrodermal activity (EDA), and eye movements—with advanced machine learning (ML) techniques, offers a promising approach to automate and objectify mental health assessments. A systematic review was conducted to explore recent advances in the early detection of anxiety and stress by combining physiological signals and ML methods. To assess methodological quality, a specific analysis framework was designed for the 113 studies included, which identified significant deficiencies in the literature. This highlights the urgent need to adopt standardized reporting guidelines in this field. The role of these technologies in feature extraction, classification, and predictive modeling was analyzed, also addressing critical challenges related to data quality, model interpretability, and the influence of intersectional factors like gender and age. Ethical and privacy considerations in the current research were also included. Finally, potential avenues for future research were summarized, highlighting the potential of ML technologies for early detection and proactive intervention in mental disorders. Full article

Background: Human papillomavirus (HPV)-associated oral and oropharyngeal squamous cell carcinomas have risen dramatically in incidence over recent decades. Yet, unlike cervical neoplasia, there is no established screening paradigm for HPV-driven oropharyngeal dysplasia, as precursor lesions are often occult and are not easily accessible for examination. This drives an urgent need for non-invasive biomarkers to enable early detection, risk stratification, and timely intervention. Objective of this review is to highlight advances in liquid biopsy modalities, specifically saliva- and blood-based biomarkers—in the context of HPV-driven oral carcinogenesis—and to evaluate their utility in early cancer detection, prognostic, post-treatment surveillance, and recurrence monitoring. Methods: We performed a narrative review of PubMed-indexed studies (2015–2025) focusing on HPV-positive oral and oropharyngeal squamous cell carcinomas. and liquid biopsy analytes. Key sources were high-impact original studies and meta-analyses from 2020–2025 examining circulating tumor DNA (ctDNA), viral nucleic acids, circulating tumor cells (CTCs), extracellular vesicles (EVs), and related biomarkers in saliva and blood. Reported data on assay performance, biases, and validation were reviewed to highlight how oral cancer findings align with trends seen in other solid tumors. Results: In reviewing recent studies (2015–2025), we found consistent evidence that saliva best captures locoregional tumor signals while plasma circulating tumor HPV DNA (ctHPV DNA) reflects systemic disease, and that using both matrices improves detection over either alone. Dual-fluid testing will potentially enable earlier identification of molecular residual disease with clinically meaningful lead time before radiographic recurrence, supporting risk-adapted surveillance. Overall, literature favors standardized pre-analytics and combined saliva plus plasma workflows to enhance early detection and follow-up in HPV-positive oral and oropharyngeal squamous cell carcinomas. Conclusions: Liquid biopsy approaches offer promising tools for the early, non-invasive detection and real-time monitoring of HPV-associated oral cancers. Realizing their full clinical potential will require robust prospective validation and standardization of pre-analytical protocols. Integrating salivary and blood biomarkers into tailored surveillance programs may further support earlier intervention and improved patient outcomes, while potentially reducing reliance on unnecessary invasive procedures. Full article

Background/Objectives: β-Carotene oxygenase-2 (BCO2) is a mitochondrial carotenoid-cleaving enzyme expressed in multiple tissues, including the lungs. While BCO2 regulates carotenoid handling, its role in shaping pulmonary immune architecture and antiviral responses is unknown. We hypothesized that BCO2 deficiency reprograms epithelial–innate circuits and alters antiviral outcomes. Methods: BCO2-knockout (KO) and C57BL/6J wild-type (WT) mice underwent lung single-cell RNA sequencing (scRNA-seq), immunoblotting, and intranasal SARS-CoV-2 challenge to assess cell-type heterogeneity, pathway programs (by gene set variation analysis, GSVA), and antiviral responses. Results: scRNA-seq resolved 14 major lung cell populations with cell-type-specific pathway shifts. Compared with WT, BCO2 KO lungs showed increased conventional dendritic cells and natural killer (NK) cells, with reductions in macrophages, B cells, and endothelial cells. In KO alveolar type II cells, GSVA indicated a stress-adapted metabolic program. Ciliated epithelium exhibited vitamin-K-responsive and axoneme-remodeling signatures with attenuated glucocorticoid and very-low-density lipoprotein remodeling. Innate lymphoid type 2 cells favored fatty acid oxidation and chromatin dynamics with reduced mitochondrial activity. NK cells were biased toward constitutive chemokine/cytokine secretion and counter-inflammatory signaling. Immunoblotting confirmed the elevated level of interferon regulatory factor-3 protein in BCO2-KO lungs. Functionally, BCO2-KO mice had improved outcomes after intranasal SARS-CoV-2 exposure. Conclusions: Loss of BCO2 reconfigures the pulmonary immune landscape and enhances antiviral responsiveness in mice. These findings identify BCO2 as a nutrient-linked enzyme with immunomodulatory impact and highlight cell-state changes as candidate mechanisms for improved antiviral tolerance. Full article

Transformer-based models have accelerated EEG motor imagery (MI) decoding by using self-attention to capture long-range temporal structures while complementing spatial inductive biases. This systematic survey of Scopus-indexed works from 2020 to 2025 indicates that reported advances are concentrated in offline, protocol-heterogeneous settings; inconsistent preprocessing, non-standard data splits, and sparse efficiency frequently reporting cloud claims of generalization and real-time suitability. Under session- and subject-aware evaluation on the BCIC IV 2a/2b dataset, typical performance clusters are in the high-80% range for binary MI and the mid-70% range for multi-class tasks with gains of roughly 5–10 percentage points achieved by strong hybrids (CNN/TCN–Transformer; hierarchical attention) rather than by extreme figures often driven by leakage-prone protocols. In parallel, transformer-driven denoising—particularly diffusion–transformer hybrids—yields strong signal-level metrics but remains weakly linked to task benefit; denoise → decode validation is rarely standardized despite being the most relevant proxy when artifact-free ground truth is unavailable. Three priorities emerge for translation: protocol discipline (fixed train/test partitions, transparent preprocessing, mandatory reporting of parameters, FLOPs, per-trial latency, and acquisition-to-feedback delay); task relevance (shared denoise → decode benchmarks for MI and related paradigms); and adaptivity at scale (self-supervised pretraining on heterogeneous EEG corpora and resource-aware co-optimization of preprocessing and hybrid transformer topologies). Evidence from subject-adjusting evolutionary pipelines that jointly tune preprocessing, attention depth, and CNN–Transformer fusion demonstrates reproducible inter-subject gains over established baselines under controlled protocols. Implementing these practices positions transformer-driven BCIs to move beyond inflated offline estimates toward reliable, real-time neurointerfaces with concrete clinical and assistive relevance. Full article

Osteoarthritis (OA) lacks disease-modifying therapies, in part because key features of the joint microenvironment remain underappreciated. One such feature is localized acidosis, characterized by sustained reductions in extracellular pH within the cartilage, meniscus, and the osteochondral interface despite near-neutral bulk synovial fluid. We synthesize current evidence on the origins, sensing, and consequences of joint acidosis in OA. Metabolic drivers include hypoxia-biased glycolysis in avascular cartilage, cytokine-driven reprogramming in the synovium, and limits in proton/lactate extrusion (e.g., monocarboxylate transporters (MCTs)), with additional contributions from fixed-charge matrix chemistry and osteoclast-mediated acidification at the osteochondral junction. Acidic niches shift proteolysis toward cathepsins, suppress anabolic control, and trigger chondrocyte stress responses (calcium overload, autophagy, senescence, apoptosis). In the nociceptive axis, protons engage ASIC3 and sensitize TRPV1, linking acidity to pain. Joint cells detect pH through two complementary sensor classes: proton-sensing GPCRs (GPR4, GPR65/TDAG8, GPR68/OGR1, GPR132/G2A), which couple to G_s, G_q/11, and G_12/13 pathways converging on MAPK, NF-κB, CREB, and RhoA/ROCK; and proton-gated ion channels (ASIC1a/3, TRPV1), which convert acidity into electrical and Ca²⁺ signals. Therapeutic implications include inhibition of acid-enabled proteases (e.g., cathepsin K), pharmacologic modulation of pH-sensing receptors (with emerging interest in GPR68 and GPR4), ASIC/TRPV1-targeted analgesia, metabolic control of lactate generation, and pH-responsive intra-articular delivery systems. We outline research priorities for pH-aware clinical phenotyping and imaging, cell-type-resolved signaling maps, and targeted interventions in ‘acidotic OA’ endotypes. Framing acidosis as an actionable component of OA pathogenesis provides a coherent basis for mechanism-anchored, locality-specific disease modification. Full article

Background/Objectives: κ-Opioid receptors have emerged as promising targets for novel analgesic strategies, offering the potential to relieve pain without the adverse effects commonly associated with μ-opioid receptor activation, such as respiratory depression, tolerance, and addiction. This review focuses on recent advances in understanding KOR-mediated pain modulation and aims to evaluate the therapeutic potential of KOR agonists in addressing the limitations of current opioid-based treatments. Methods: This review synthesizes evidence from comprehensive preclinical studies investigating the effects of KOR agonists on central pain pathways, including modulation of neurotransmitter release and attenuation of ascending nociceptive signaling. In addition, emerging clinical trial data on KOR-selective compounds will be evaluated, together with recent advances in biased agonism and region-specific receptor signaling, to guide the development of next-generation analgesics. Results: Preclinical studies demonstrate robust antinociceptive effects of KOR agonists, while early clinical trials indicate that several KOR-selective compounds effectively reduce pain symptoms. Advances in biased agonism and targeted receptor signaling suggest the potential to achieve analgesia with reduced dysphoria and sedation. Conclusions: KOR-targeted therapies show significant translational potential for pain management. The integration of preclinical and clinical evidence supports the development of next-generation KOR agonists that could provide effective analgesia while minimizing the adverse effects associated with conventional opioids. Full article

Mutations may cause changes in the structure and function of immune-related proteins, thereby affecting the operation of the immune system. Deep mutational scanning combines saturation mutagenesis, functional selection, and high-throughput sequencing to evaluate the effects of mutations on a large scale and with high resolution. By systematically and comprehensively analyzing the impact of mutations on the functions of immune-related proteins, the immune response mechanism can be better understood. However, each stage in deep mutation scanning has its limits, and the approach remains constrained in several ways. These include data and selection biases that affect the robustness of effect estimates, insufficient library coverage and editability leading to uneven representation of sites and alleles, system-induced biased signals that deviate phenotypes from their true physiological state, and imperfect models and statistical processing that limit extrapolation capabilities. Therefore, this technology still needs further development. Herein, we summarize the principles and methods of deep mutational scanning and discuss its application in immunological research. The aim is to provide insights into the broader application prospects of deep mutational scanning technology in immunology. Full article