Search Results (336)

Clostridioides difficile infection (CDI) remains a major cause of healthcare-associated morbidity and mortality, particularly among older adults and patients with recent healthcare exposure, underscoring the need for early risk stratification and accurate prognostic assessment. This retrospective observational study evaluated clinical, laboratory, and therapeutic factors associated with disease severity and in-hospital mortality, and assessed the predictive performance of the ATLAS score and the Charlson comorbidity index. A total of 101 adult inpatients with laboratory-confirmed CDI admitted to a Portuguese tertiary care hospital were included. Data were extracted from clinical records and analysed using comparative statistics, multivariable logistic regression, and Kaplan–Meier survival analysis. Advanced age, elevated white blood cell count, renal dysfunction, and prior exposure to multiple antibiotic classes were independently associated with increased disease severity and mortality. The ATLAS score demonstrated good discriminative ability, particularly for short-term mortality, and showed higher sensitivity compared with the Charlson comorbidity index. These findings provide additional evidence on clinical and laboratory factors associated with severe CDI and in-hospital mortality, while supporting the utility of the ATLAS score as a practical tool for early risk stratification in hospitalised patients. Full article

Selecting a source language for zero-shot cross-lingual transfer is typically done by intuition or by defaulting to English, despite large performance differences across language pairs. We study whether linguistic similarity can predict transfer performance and support principled source-language selection. We introduce quantified WALS (qWALS), a typology-based similarity metric derived from features in the World Atlas of Language Structures, and evaluate it against existing similarity baselines. Validation uses three complementary signals: computational similarity scores, zero-shot transfer performance of multilingual transformers (mBERT and XLM-R) on four NLP tasks (dependency parsing, named entity recognition, sentiment analysis, and abusive language identification) across eight languages, and an expert-linguist similarity survey. Across tasks and models, higher linguistic similarity is associated with better transfer, and the survey provides independent support for the computational metrics. Full article

To assess the therapeutic relevance of BRIP1 in gastric cancer (GC), we examine its functional role in conferring resistance to anoikis within GC cells and elucidate the oncogenic signaling pathways modulated by BRIP1. By integrating the Cancer Genome Atlas (TCGA) and Gene Set Enrichment Analysis (GSEA) databases with Least Absolute Shrinkage and Selection Operator (LASSO) regression, a novel risk score to stratify GC patients based on prognosis was generated, and a significantly differentially expressed gene, BRIP1, was validated through reverse transcription quantitative polymerase chain reaction (RT-qPCR). Protein expression associated with apoptosis, cell cycle, and epithelial-mesenchymal transformation (EMT) was quantified via RT-qPCR and Western blot (WB). 5-Ethynyl-2′-deoxyuridine (EdU) and cell counting kit-8 (CCK-8) assays were conducted to quantify proliferative activity. The protein level in axillary tumor tissues of nude mice was detected by immunohistochemistry (IHC). We established an eight-gene anoikis-related prognostic risk assessment model (DUSP1, VCAN, P3H2, TXNIP, BRIP1, FGD6, GPX3, and RLN2) for GC. Multivariate Cox regression confirmed the risk score as an independent prognostic factor. Among these genes, BRIP1 showed significant differential expression between tumor and normal tissues, as well as normal gastric mucosal epithelial cells and GC cells. Mechanistically, BRIP1 conferred anoikis resistance to GC cells by suppressing the generation of reactive oxygen species (ROS). We found that the PI3K inhibitor LY294002 counteracted BRIP1-driven oncogenic effects, which was evidenced by restored expression of key regulators governing apoptosis, cell cycle progression, and EMT, in addition to suppressed proliferation in GC cells. BRIP1 is postulated to function upstream of the PI3K/Akt signaling cascade. This study establishes a risk scoring model and identifies BRIP1 as a potential prognostic marker for GC. Full article

Advanced Persistent Threats (APTs) are characterized by stealth, infrequency, and long cycles, evading traditional security to endanger critical infrastructure. Complex semantic links between system entities can be accurately modeled using representation learning techniques based on heterogeneous provenance graphs, providing a novel method for uncovering hidden APT attack chains. However, in large-scale practical implementations, this approach still faces three major challenges: combinatorial explosion of long-range meta-paths, loss of semantic evolution during graph compression, and high computational overhead for dynamic environments. To address these, we propose APT-LMSPS, a detection system leveraging Long-Range Meta-path Progressive Sampling Search (LMSPS). The LMSPS algorithm uses dynamic pruning and semantic contribution assessment to convert meta-path combination explosion into constant-scale computation, accurately modeling long-range dependencies. Second, the Maintaining Global Semantics (MGS) approach intelligently filters events by tracking node semantic state changes, achieving an 8:1 compression ratio while preserving over 90% of critical pathways’ semantic integrity. Lastly, the meta-path encoding database uses a caching approach to avoid repeated encoding, doubling encoding effectiveness and enabling efficient, accurate, system-wide APT detection in large-scale scenarios. Evaluated on DARPA, StreamSpot, and ATLAS datasets, APT-LMSPS maintains competitive accuracy (F1-score ≥ 0.98) and improves long-range processing efficiency by an order of magnitude over baselines. Full article

The integration of machine learning (ML) algorithms in oncology creates a new path for prognosis, early diagnosis, prevention, and treatment customization. However, large-scale clinical implementation is difficult due to the lack of standardized assessments and the variation in reported performance. A systematic review of the most recent research on ML applications in oncology (1 January 2020–31 December 2025) was conducted. The databases employed are Web of Science, Scopus, and PubMed. Filters applied for open-access articles that were simultaneously indexed and had numerical data in the abstract. From an initial of 13,292 articles, successive selection according to the PRISMA diagram resulted in a final set of 1364 studies. These were analyzed from four perspectives: the types of cancer investigated, the characteristics of the datasets (reproducibility and generalizability), the ML models used, and the performance achieved (accuracy, precision, recall, F1-score, and AUC). There is high interest in breast cancer (350 articles), colorectal cancer (337 articles), and lung cancer (151 articles), with frequent use of the databases The Cancer Genome Atlas (133 studies), Gene Expression Omnibus (94 studies), and Surveillance, Epidemiology, and End Results (72 studies). The Random Forest model proved to be predominant due to its tolerance for incomplete data. Reported performance varies considerably between cancer types and even within the same type. This analysis demonstrates the potential of ML methods for deciphering genomic alterations and supports the development of integrated personalized medicine approaches in oncology. Full article

Background/Objectives: Hepatocellular carcinoma (HCC) remains a leading cause of cancer-related mortality worldwide, where accurate liver and tumor segmentation from magnetic resonance imaging (MRI) is essential for diagnosis, treatment planning, and disease monitoring. Despite recent advances, MRI-based segmentation remains challenging due to data heterogeneity and limited annotated datasets. This study aims to systematically compare convolutional, transformer-based, and detection-based deep learning approaches for liver and HCC segmentation using contrast-enhanced MRI. Methods: A comprehensive evaluation was conducted on the ATLAS MRI dataset, including 2D- and 3D-CNN, transformer-based architectures, and single-stage YOLO-based segmentation frameworks. All models were trained using consistent preprocessing, patient-level data splits, and standardized evaluation metrics, including Dice coefficient, Intersection over Union (IoU), precision, recall, and F1-score. Results: Volumetric convolutional models achieved the highest segmentation accuracy, with the 3D nnU-Net yielding superior performance for both liver (Dice: 0.946) and tumor (Dice: 0.892) segmentation. Transformer-based models demonstrated competitive results, particularly in capturing global contextual information and improving boundary delineation, while YOLO-based approaches provided balanced accuracy with substantially reduced computational cost. Conclusions: The findings confirm that volumetric CNNs remain the most accurate solution for MRI-based liver and HCC segmentation, whereas transformer- and YOLO-based frameworks offer complementary advantages for specific clinical and resource-constrained scenarios. Full article

Gliomas are the most prevalent primary malignant neoplasms of the central nervous system, distinguished by their high recurrence rates and poor prognosis. Aerobic glycolysis in tumors generates excess lactate, which promotes lactylation, a post-translational modification (PTM). Although accumulating evidence implicates lactylation in glioma initiation and progression, previous lactylation-focused prognostic studies lacked single-cell resolution and broad validation, limiting their generalizability and clinical relevance. Single-cell and bulk RNA sequencing (RNA-seq) data were integrated to identify lactylation-enriched tumor cell populations and derive candidate genes. A risk model was developed using univariate Cox regression and the Least Absolute Shrinkage and Selection Operator (LASSO), and its predictive performance was validated in independent cohorts from the China Glioma Genome Atlas (CGGA). To improve clinical applicability, a nomogram integrating the risk score incorporating key clinical variables was constructed and externally validated. The risk groups showed distinct immune microenvironment profiles and differential drug sensitivity patterns. In this study, we established and validated a lactylation-related gene signature, with the derived risk score serving as a reliable prognostic biomarker for glioma. Furthermore, the model not only predicts overall survival (OS) but also exhibits the potential to inform drug selection and stratify patients for more precise and personalized therapeutic interventions. Full article

Glioblastoma (GBM) shows extensive epigenetic heterogeneity. In IDH-wildtype (IDH-WT) GBM, promoter DNA methylation may regulate lineage programs influencing tumor evolution and prognosis; here, we systematically profiled promoter-level methylation dynamics across longitudinal tumors. Genome-wide DNA methylation data were obtained from the publicly available Gene Expression Omnibus (GEO; GSE279073) dataset, comprising a longitudinal cohort of 226 IDH-wildtype glioblastomas profiled on the Illumina Infinium EPIC 850K array across primary and recurrent stages at the University of California, San Francisco. From 333 Gene Ontology gliogenesis-annotated genes (GO:0042063), a 48-gene promoter panel was derived, with ≥2 probes per gene. Promoter methylation was summarized as the median β-value and tested using one-sample Wilcoxon with FDR correction. Functional enrichment, longitudinal variation, and patient-level methylation burden were assessed. Validation analyses were performed using independent IDH-wildtype GBM datasets from The Cancer Genome Atlas (RNA-seq and 450K methylation; n = 347). Promoter hypomethylation predominated across all stages, with 25 genes consistently hypomethylated and 7 hypermethylated. Functional enrichment highlighted gliogenesis, glial cell differentiation, neurogenesis, and Notch-related signaling. In TCGA, promoter methylation inversely correlated with expression for 11 of 33 genes (FDR < 0.05). An Expression Score contrasting hypomethylated and hypermethylated genes was positively associated with improved overall survival, where higher scores predicted better outcome (HR = 0.87, p = 0.016; Q4 vs. Q1 HR = 0.68, p = 0.025), and a complementary Methylation Score showed that higher promoter hypermethylation predicted poorer outcome (HR = 1.73, p < 0.001). CNTN2 and TSPAN2 were adverse prognostic genes (FDR < 0.05). The Expression Score was highest in Proneural tumors and lowest in Mesenchymal tumors (p < 0.001), reflecting a proneural-like state associated with better prognosis. Promoter methylation within gliogenesis genes defines a stable yet prognostically informative epigenetic signature in IDH-WT GBM. Hypomethylation promotes transcriptional activation and a favorable outcome, whereas hypermethylation represses lineage programs and predicts poorer survival. Full article

Cellular senescence, characterized by permanent cell cycle arrest, significantly influences cancer development, immune regulation, and progression. However, the precise mechanisms by which senescence contributes to colorectal cancer prognosis remain to be fully elucidated. By integrating expression profiles of senescence-related and prognostic genes in colon adenocarcinoma (COAD) patients, we formulated and confirmed a nine-gene cellular senescence-related signature (CSRS) that integrates senescence-associated and prognosis-predictive genes using data from the CellAge, The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO). A cell senescence-related prognostic formula was developed as follows: CSRS = (CASP2 × 0.2098) + (CDKN2A × 0.1196) + (FOXD1 × 0.1472) + (ING5 × 0.3723) + (OXTR × 0.0786) + (PHGDH × 0.1408) + (SERPINE1 × 0.1127) + (SNAI1 × 0.1034) + (LIMK1 × 0.0747). In a multivariate Cox proportional hazards model, the CSRS score, age and TNM stage were all identified as significant independent indicators for overall survival, affirming their prognostic value in colorectal cancer. The CSRS-high group exhibited significantly up-regulated senescence-associated secretory phenotype (SASP) and immune cell infiltration, whereas the CSRS-low group showed an apparent better response to immune-checkpoint inhibitor therapy. Our findings suggest CSRS score and its constituent genes represent potential biomarkers for prognosis and immunotherapeutic benefit in COAD patients. Extending this nine-gene set into a broader senescence-associated panel should be a next step toward delivering truly individualized treatment plans. Full article

Background and Objective: Patients with stage III clear cell renal cell carcinoma (ccRCC) have a high risk for disease recurrence post-nephrectomy. To mitigate overtreatment, there is a pressing need to determine who benefits from immune checkpoint inhibition (ICI) around the time of surgical resection. We performed digital spatial analysis of both gene and protein expression in stage III ccRCC tumors, some of which had preoperative ICI exposure. Methods: Nephrectomy specimens from stage III ccRCC patients were analyzed using the Nanostring GeoMx Digital Spatial Profiler. Differential expression analysis was performed and validated using NCT02210117 trial data to identify genes associated with both ICI and clinical response. A gene score was then generated to predict overall survival in patients from The Cancer Genome Atlas (TCGA). Key Findings and Limitations: In a small cohort of 19 patients, RNA expression significantly differed based on preoperative ICI exposure and recurrence status—CD8+ effector and central-memory T-cell signatures were less prevalent in the treatment-naïve with recurrence group. Three out of four patients with preoperative immune checkpoint inhibition recurred. External validation yielded a four-gene set (GZMK, GZMA, ITGAL, and IL7R), where higher expression levels predicted better overall survival in the TCGA cohort (p = 0.005). Conclusions and Clinical Implications: Preoperative ICI favorably altered the tumor microenvironment to resemble that of treatment-naïve patients without recurrence but did not translate to improved survival. Upon external validation, the genes GZMK, GZMA, ITGAL, and IL7R were modifiable with ICI and associated with improved overall survival. Further investigation is needed to assess if patients with low baseline expression of these genes may benefit from ICI around the time of surgery. Full article

Background: Head and neck squamous cell carcinoma remains a highly aggressive malignancy with limited predictive biomarkers for prognosis and radiotherapy response. Increasing evidence indicates that pseudogenes are functionally active regulators of cancer biology, yet their clinical relevance in HNSCC is poorly defined. Methods: Using transcriptomic and clinical data from The Cancer Genome Atlas, we analyzed the expression and clinical significance of two pseudogenes, ANXA2P2 and PA2G4P4, in HNSCC. Associations with clinicopathological features, HPV status, tumor subtypes, survival, genomic instability, radiotherapy response, and immune landscape were assessed using bioinformatic tools. Results: Both pseudogenes were significantly upregulated in HNSCC compared to normal tissues. Higher expression levels correlated with adverse clinicopathological features, increased tumor proliferation and wound-healing capacity, and unfavorable TCGA molecular subtypes. High ANXA2P2 and PA2G4P4 expression was associated with reduced overall survival, while their combined low-expression signature identified patients with significantly improved overall and disease-free survival. Notably, lower expression of both pseudogenes was observed in patients responding to radiotherapy, whereas higher expression was linked to genomic instability parameters and enrichment of oncogenic pathways, including MYC, PI3K/AKT/mTOR, cell cycle regulation, and DNA repair. ANXA2P2 expression differed significantly by HPV status, showing reduced levels in HPV-positive tumors. Furthermore, pseudogene expression stratified distinct immune profiles, including immune subtypes, stromal and immune scores, and specific immune cell populations. Conclusions:ANXA2P2 and PA2G4P4 are clinically relevant pseudogenes associated with tumor aggressiveness, immune modulation, and radiotherapy response in HNSCC. These findings support their potential utility as prognostic and predictive biomarkers and provide a rationale for further functional validation in experimental models. Full article

Background/Objectives: Head and neck squamous cell carcinoma (HNSCC) is a biologically heterogeneous malignancy with poor outcomes in advanced disease. Increasing evidence indicates that the tumor microenvironment, particularly cancer-associated fibroblasts (CAFs), plays an important role in tumor progression and immune regulation. However, the diversity of CAF subsets and their clinical relevance in HNSCC remain incompletely understood. This study aimed to characterize CAF heterogeneity and assess the prognostic significance of CAF subset-specific transcriptional programs. Methods: Single-cell RNA sequencing data from HNSCC tumors were analyzed to identify CAF subsets based on differentially expressed genes. CAF subset-specific gene signatures were used to construct prognostic risk models for overall survival (OS) and progression-free survival (PFS) in The Cancer Genome Atlas HNSCC cohort, with validation in an independent dataset. CAF-driven prognostic groups were defined, and their immune landscapes and biological pathways were evaluated. Bulk RNA sequencing of primary CAF cultures was performed for validation. Results: Six CAF subsets were identified, including myofibroblastic (myCAF), inflammatory (iCAF), antigen-presenting, and extracellular matrix-related CAFs. Risk scores derived from inflammatory CAF subsets consistently predicted shorter OS across independent cohorts, whereas PFS prediction showed greater cohort dependency. CAF-based stratification identified patient subgroups with distinct immune profiles and pathway enrichment patterns. These results were supported by validation analyses and by bulk RNA sequencing of primary CAFs, demonstrating preservation of myCAF- and iCAF-like transcriptional programs ex vivo. Conclusions: CAF heterogeneity has important prognostic and immunological implications in HNSCC. Inflammatory CAF-related transcriptional programs represent robust markers of patient survival and may complement tumor-intrinsic biomarkers. Full article

Smart cities increasingly reuse existing Wi-Fi infrastructure to sense crowding, but many smart-campus tools still fail to support routine, day-to-day decisions. A short-horizon field feasibility study was conducted to prototype a low-maintenance, prefix-based early-warning rule that turns anonymized campus Wi-Fi access-point counts into an interpretable lunchtime crowd signal. Daily 7-min access-point profiles from five university canteens (11:00–14:00) were aggregated, winsorized, smoothed, and row-z-scored, then clustered into demand-shape typologies using k-means++. Two typologies were obtained (Early Peak and Late Shift), and a cosine-similarity atlas was frozen. At 11:28, the five-bin occupancy prefix was compared to typology centroids, and an Early Peak badge was issued when similarity to the Early Peak centroid exceeded a preset threshold. On held-out days, the Early Peak typology could be identified at 11:28 with coverage of 0.73 and agreement of 0.86 relative to end-of-day labels. In 20 matched canteen-weekday pairs, badge days were associated with a Hodges–Lehmann median reduction of 0.193 standard-deviation units in peak crowding (≈9% lower). Given the short (3-week) horizon and limited hold-out window, results are presented as feasibility evidence and motivate a larger controlled evaluation. Simple, interpretable rules built on existing Wi-Fi telemetry were shown to be deployable as a feasibility-level decision aid on a smart campus, while broader smart-city transferability should be validated through longer-horizon controlled evaluations. Full article

Background: Lung adenocarcinoma (LUAD) exhibits pronounced cellular and molecular heterogeneity that shapes tumor progression and therapeutic response. Although nucleotide metabolism is essential for sustaining tumor proliferation and coordinating immune interactions, its single-cell heterogeneity and clinical implications remain incompletely defined. Methods: We integrated a publicly available scRNA-seq dataset derived from independent LUAD patients to construct a comprehensive LUAD cellular atlas, identified malignant epithelial cells using inferCNV, and reconstructed differentiation trajectories via Monocle2. Cell–cell communication patterns under distinct nucleotide metabolic states were assessed using CellChat. A nucleotide metabolism-related signature (NMRS) was subsequently developed across TCGA-LUAD and multiple GEO cohorts using 101 combinations of machine learning algorithms. Its prognostic and immunological predictive value was systematically evaluated. The functional relevance of the key gene ENO1 was further verified through pan-cancer analyses and in vitro experiments. Results: We identified substantial nucleotide metabolic heterogeneity within malignant epithelial cells, closely linked to elevated proliferative activity, glycolytic activation, and increased CNV burden. Pseudotime analysis showed that epithelial cells gradually acquire enhanced immune-modulatory and complement-related functions along their differentiation continuum. High-metabolism epithelial cells exhibited stronger outgoing communication—particularly via MIF, CDH5, and MHC-II pathways—highlighting their potential role in shaping an immunosuppressive microenvironment. The NMRS built from metabolism-related genes provided robust prognostic stratification across multiple cohorts and surpassed conventional clinical parameters. Immune profiling revealed that high-NMRS tumors displayed increased T-cell dysfunction, stronger exclusion, higher TIDE scores, and lower IPS, suggesting poorer responses to immune checkpoint blockade. ENO1, markedly upregulated in high-NMRS tumors and functioning as a risk factor in several cancer types, was experimentally shown to promote invasion in LUAD cell lines. Conclusions: This study delineates the profound impact of nucleotide metabolic reprogramming on epithelial cell states, immune ecology, and malignant evolution in LUAD. The NMRS provides a robust predictor of prognosis and immunotherapy response across cohorts, while ENO1 emerges as a pivotal metabolic–immune mediator and promising therapeutic target. Full article

Background: Vesicoureteral reflux (VUR) contributes significantly to recurrent childhood urinary tract infections and renal scarring, yet predicting which patients will develop adverse outcomes or benefit from specific investigations or treatments remains challenging. Numerous prognostic tools have been proposed, but none have achieved widespread adoption. Methods: A comprehensive search of the literature available on MEDLINE, PUBMED, Embase, Emcare, CINAHL, and Google Scholar was performed to identify combinations of factors, scoring systems, ratios, models, and tools relating to VUR. This included predicting the spontaneous resolution of established vesicoureteral reflux, the risk of breakthrough urinary tract infections (UTIs), and guiding clinical decision making regarding the need for VCUG in patients with UTIs, continuous antibiotic prophylaxis (CAP), or surgical intervention in patients with confirmed VUR. Articles were included if they either described or validated a predictive tool that was designed to aid clinical decision making in patients with either suspected or confirmed VUR with regards to investigation or management strategies. All the studies included were then analysed, and the predictive tools have been summarised in a narrative format. Results: Seventeen predictive tools developed over thirty-nine years were identified: six predicting spontaneous resolution, four predicting breakthrough urinary tract infection (BTUTI) on CAP, two determining which children benefit from CAP, and five estimating the probability of VUR or high-grade VUR after a first febrile UTI. Approaches ranged from radiological ratios to multifactorial clinical–radiological scores and machine-learning models. Only five tools had any external validation, and none demonstrated sufficient reliability for universal clinical use. Significant heterogeneity in design, imaging interpretation, inclusion criteria, and outcome definitions limited comparison and wider applicability. Conclusions: This atlas provides the first consolidated overview of prognostic tools in paediatric VUR. Future development should prioritise multicentre, prospectively validated models that integrate established clinical and radiological predictors with transparent computational methods to create practical, generalisable risk-stratification frameworks for routine care. Full article