Search Results (287)

Despite significant advancements in Artificial Intelligence, its widespread adoption in the clinical domain remains restricted due to the inherent complexity, fragmented nature, and diversity of healthcare systems. Each healthcare provider has unique data, clinical guidelines, data availability, system architectures, heterogeneity, and distribution. These challenges hinder the application of Clinical Decision Support Systems because of a limited understanding of how existing systems can be effectively redeployed across different healthcare providers. Redeployment is needed because it enables the reuse of existing knowledge, maximizes reusability, and avoids code duplication, thereby reducing the costs, effort, and time required to develop the Clinical Decision Support System from scratch. In addition, it ensures faster deployment and wider accessibility in the case of resource-constrained healthcare providers. An essential for redeployment is to identify the possible situations in which variables differ between two dynamic environments. To address this gap, we propose a structured multi-dimensional framework that systematically analyzes the potential differences between the variables. To represent the output of differences across dimensions based on variables in a systematic, machine-readable manner, we proposed a conceptual model, “VarDiff”, and a decision matrix of possible outcomes across five differential dimensions. This conceptual model provides a systematic, structural, and logical representation of a multidimensional framework for identifying differences among variables across data ecosystems. It formalizes variable characteristics in terms of semantic entities to observe differences among variables. The adaptation categories help identify the specific adaptation type, enabling the selection of relevant adaptation strategies in the “Mutator” component. Full article

Welding is a critical joining process in civil and transportation infrastructure, enabling the fabrication of complex steel structural systems used in bridges, buildings, and other essential infrastructures. Despite strict adherence to established welding codes and standards, such as AWS D1.1 and AASHTO/AWS D1.5, welding flaws and service-induced defects can occur in welded components. Cause of defects and their structural impact, along with detection, sizing, and localization of these anomalies and flaws, are crucial for adequate maintenance, repair, or replacement planning without compromising the functionality of in-service components. Among available NDT techniques, ultrasonic testing (UT) remains one of the most widely adopted methods of weld inspection due to its depth of penetration, sensitivity to internal defects, and suitability for field deployment. Recent advancements in ultrasonic technologies, particularly Phased Array Ultrasonic Testing (PAUT), along with its emerging approaches such as Full Matrix Capture (FMC) and the Total Focusing Method (TFM), have significantly enhanced inspection accuracy, repeatability, and interpretability. These techniques enable flexile beam steering, multi-angle interrogation, and improved imaging of complex geometries. This paper presents a comprehensive review of PAUT for the inspection of welded steel infrastructure adhering to the recommendations and requirements of the relevant codes and standards, synthesizing the current literature on PAUT principles, wave modes, probe configurations, and data acquisition strategies. Emphasis is placed on the practical implementation of PAUT in civil infrastructure inspection, its advantages over conventional NDT methods, and its potential to support informed decisions related to quality acceptance, repair, and long-term maintenance planning. This paper concludes by identifying current challenges and future research directions for advanced ultrasonic inspection of welded steel structures. Full article

Building code waiver assessments in Singapore remain largely discretionary, relying on case officers’ subjective judgement with limited decision-support tooling. This study presents the first machine learning framework for predicting building code waiver outcomes, trained on 197 historically decided cases from the Building and Construction Authority (BCA) across five waiver categories: barrier-free accessibility (n = 45), ventilation (n = 61), staircase design (n = 37), safety provisions (n = 30), and structural modifications (n = 24), spanning 2021 to 2023. Fourteen engineered features, including documentation completeness, technical justification quality, and compliance history, were extracted through domain-expert annotation. Four models were evaluated: L2-regularised logistic regression, random forest, gradient boosting (XGBoost 2.0.1), and a weighted ensemble. The ensemble achieved the highest predictive accuracy of 83.7% (95% CI: 79.2–88.1%) with an area under the receiver operating characteristic curve (AUC) of 0.891 (95% CI: 0.854–0.928), significantly outperforming all individual models (McNemar’s test, p < 0.05). SHAP analysis revealed that documentation completeness and technical justification quality collectively account for 55% of prediction variance. A companion five-by-five risk assessment matrix, combining predicted rejection probability with consequence severity, stratified cases into actionable risk tiers correlating with observed approval rates ranging from 90.3% (very low risk) to 10.0% (very high risk; Spearman rho = −0.71, p < 0.001). Performance varied across waiver categories: ventilation waivers achieved the highest balanced accuracy (87.1%) while safety waivers proved most challenging (balanced accuracy 64.3%, sensitivity 40.0%). The framework offers a transparent, data-driven decision-support complement to regulatory judgement, learning patterns from historically decided applications within the 2021–2023 BCA context, and demonstrates feasibility for integration into Singapore’s Corenet X digital building submission platform. These five waiver categories serve as domain stratification variables. The machine learning target variable is the binary regulatory outcome: Approved (46.2% of cases) or Rejected (53.8%). Full article

Chronic lymphocytic leukemia (CLL) is uncommon in Asia, and longitudinal genomic data from Asian cohorts are limited. We conducted serial whole-exome sequencing (WES) in a multicenter Korean cohort of newly diagnosed, elderly CLL treated with chlorambucil–obinutuzumab to evaluate mutational heterogeneity and clonal hematopoiesis of indeterminate potential (CHIP) during treatment and follow-up. Tumor-only variants were filtered, restricted to nonsynonymous or loss-of-function coding/splice-site mutations, and summarized as a binary patient-by-gene matrix for principal component analysis (PCA), trajectory analysis, and k-means clustering. CHIP was defined as ≥1 qualifying mutation in a prespecified CHIP gene set. Baseline PCA was more compact in patients with complete response at end of treatment, whereas partial response or progressive disease cases were more dispersed. PCA trajectories were compact and directionally consistent in complete responders, more dispersed in partial responders, and highly heterogeneous without a dominant direction in progressive disease. Clustering identified dispersed and compact clusters, and CHIP-associated mutations were enriched in the dispersed cluster (55.6% vs. 8.3%, Fisher’s exact p = 0.0086). In paired samples collected 3–5 months after end of treatment, CHIP status changed in some patients. Serial WES may provide complementary information to treatment response, although these observations require confirmation in larger cohorts. Full article

In the digital era, even the most remote communities are increasingly connected to global networks. However, a critical question persists: how can such connectivity translate into tangible economic growth and sustainable development for isolated mountainous villages? Guided by the sustainable livelihood framework, this study investigates how digital capital—specifically the use of social media to showcase a village’s natural and cultural assets—drives tourism development and improves local livelihoods. Focusing on Dazhai Village in China, a rural community that gained substantial online attention and tourism inflow through social media promotion, this research employs qualitative methods, including 17 semi-structured interviews. Data were analysed using thematic analysis and matrix coding techniques via NVivo 12 Plus. Findings reveal that the introduction of digital capital enhances village visibility, stimulates tourist interest, and initiates a development trajectory describe as “going live.” In contrast, “going alive” refers to the process of revitalizing a once abandoned, impoverished mountain village, enabling it to survive and thrive once more. However, the sustainability of this trajectory is fragile as the departure of influential digital promoters can deplete digital capital, undermining diminishing online engagement and risking renewed marginalization. To transform “going live” into “going alive,” remote communities must continuously adapt and reinforce their online presence to secure long-term stakeholders’ engagement and resilient tourism flows. An interesting finding of this study is that the village achieved regenerative tourism, whereby its environmental conditions improved as a result of tourism development. This unexpected outcome was facilitated by sustained visibility, both online and offline, which prompted residents to place greater emphasis on environmental protection. This study enriches the sustainable livelihoods framework by integrating digital capital and regenerative tourism into the understanding of livelihood assets and outcomes in remote settings. Ultimately, it underscores the transformative potential of digital capital in revitalizing “hollowed-out” villages, offering a strategic pathway for remote communities to reclaim their developmental agency and achieve sustainable rural revitalization. Full article

This study conducts a systematic literature review (SLR) of 406 peer-reviewed studies on sustainability in higher education published between 2014 and 2025. Guided by the PRISMA 2020 framework and the PICo criteria, this review identifies thematic patterns, institutional enablers, and barriers shaping sustainability integration. Data were manually screened and thematically coded using a structured extraction template. The findings reveal a conceptually active yet uneven field, with curriculum and pedagogy dominating discourse, while leadership, policy coherence, transformative learning, and global citizenship are less examined. Barriers such as institutional inertia and fragmented policies persist, but enabling factors, including digital agility, collaborative governance, and community partnerships, are attracting attention. Resilience and climate change education remain underexplored, indicating a gap between institutional strategies and sustainability goals. This review contributes by (i) identifying critical under-researched areas, (ii) refining a keyword framework to guide future inquiry, and (iii) introducing the Sustainability in Higher Education (SHE) Institutional Maturity Matrix (SHE-IMM), a conceptual model categorising institutions into foundational, transitional, and transformative stages of sustainability integration. The review received no external funding, and the authors declare there are no competing interests. Full article

Cultural heritage can strengthen urban resilience when mobilized as educational infrastructure that builds stewardship, place attachment, and civic agency. This study examines whether the Art Nouveau Path, an outdoor mobile augmented reality heritage game in Aveiro, Portugal, can function as a curriculum-aligned pathway for sustainability competences and resilience-relevant meaning-making in formal education. A curriculum translation matrix mapped eight points of interest and 36 tasks to Portuguese curriculum anchors, Education for Sustainability themes, GreenComp sustainability competences, and the Sustainable Development Goals, framing the matrix as an adoption-oriented design artefact. Empirical evidence comprised accompanying teachers’ in-field observations (T2-OBS; N = 24 across 18 sessions) and students’ post-activity survey data (S2-POST; N = 439), with open-ended reflections coded through a directed resilience-mechanism codebook (Krippendorff’s alpha = 0.91). Teachers reported high perceived value and feasibility and frequently noted enacted stewardship and placed responsibility during sessions. Students’ reflections most often linked resilience to sustainable conservation under pressure and to nature-city interconnections, whereas hazard-memory mechanisms appeared less often. Adoption-related evidence is limited to teacher feasibility reports and institutional legibility from curriculum translation, rather than confirmed institutional uptake indicators. Scaling is likely to require explicit supports for differentiation, assessment scaffolds, and routine delivery in public spaces. Full article

This systematic review analyzes the evolution of Machine Learning and Big Data applications in agriculture from 2021 to 2025, with particular emphasis on how recent technological advances facilitate the transition from precision agriculture to Intelligent Agricultural Ecosystems. A comprehensive literature search was conducted across Scopus, Web of Science, IEEE Xplore, the ACM Digital Library, SpringerLink, and MDPI, following the PRISMA 2020 guidelines. After duplicate removal and a two-stage screening process (title/abstract screening followed by full-text assessment), eligible peer-reviewed studies were systematically extracted using a structured coding matrix encompassing six analytical domains: crops, soil, weather and water, land use, animal systems, and farmer decision-making. The findings reveal a substantial increase in ML-driven agricultural analytics. Although Random Forest and Convolutional Neural Networks remain widely adopted, recent studies demonstrate a marked shift toward advanced Deep Learning architectures, integrated cloud–edge–device infrastructures, Federated Learning frameworks for privacy-preserving collaboration, Explainable AI techniques to enhance transparency, and governance-oriented mechanisms to ensure interoperability. Notwithstanding these advances, several persistent challenges remain, including limited generalizability across diverse agroclimatic contexts, the high costs associated with high-quality data annotation, the integration of heterogeneous and multimodal datasets, and infrastructural constraints related to connectivity. These developments are synthesized within the IAE conceptual framework, underscoring governance- and lifecycle-aware orchestration MLOps as a critical differentiator that transcends purely technology-centric approaches. Full article

In this paper, new fast algorithms for computing the discrete cosine transform type VI (DCT-VI) are proposed, with a special emphasis on short input sequences of three to eight samples. Fast algorithms for small discrete trigonometric transformations are directly used for efficient processing of small data sets and also serve as fundamental building blocks for constructing algorithms for larger trigonometric transforms. By exploiting the intrinsic structural properties of the DCT-VI matrices of different sizes, the proposed methods significantly reduce arithmetic complexity compared to the conventional matrix–vector multiplication approach. The paper presents a detailed mathematical formulation of the algorithms, supported by data-flow graphs that illustrate the computational structure and facilitate the precise estimation of arithmetic operations. Optimized pseudocode implementations incorporating variable reuse are also introduced to facilitate practical realization in software environments. Performance analysis demonstrates a substantial reduction in the number of multiplications (up to 66%) and a slight decrease in additions (approximately 9%) for input sizes ranging from three to eight, thereby improving the execution speed of the considering transform. The proposed algorithms are well-suited for applications in video coding, data compression, and digital signal processing, where computational efficiency is critical. Full article

Background: Triple-negative breast cancers (TNBCs) are among the most aggressive breast tumors, due not only to the absence of clinically functional biomarkers used in other molecular subtypes, but also their marked heterogeneity and pronounced migratory and invasive behavior. The search for new molecules of interest for risk prediction, diagnosis and therapy stems from the class of long non-coding RNAs (lncRNAs), which often display context-dependent (“dual”) functions and tissue specificity. Among them, lncRNA LINC01133 stands out for its dysregulation across cancer, although its molecular role in TNBC remains unclear. Methods: In the present study, we used the human TNBC cell line Hs578T to generate a cell panel comprising the parental line (Hs578T_wt), the control line (Hs578T_ctr), and the LINC01133 knockout line (Hs578T_ko). Subsequently, we performed bulk RNA-Seq to identify KO-associated Differentially Expressed Genes (DEGs) using ko_vs_ctr as the primary contrast. Functional interpretation was achieved by Over-Representation Analysis (ORA) using Gene Ontology. We then conducted a comparative patient-cohort analysis using TCGA-BRCA Basal-like/TNBC cases (TCGA/BRCA n = 1098; Basal-like/TNBC n = 199), classified with the AIMS algorithm, and evaluated concordance between KO-associated signatures and patient tumor expression patterns via trend-based analyses across the LINC01133 expression levels and associated genes. Results: A total of 265 KO-dominant DEGs were identified in Hs578T_ko, reflecting transcriptional changes consistent with tumor progression, with enrichment of pathways associated with LINC01133 knockout including cell adhesion, cell–cell interactions, epithelial–mesenchymal transition (EMT), and extracellular matrix (ECM) remodeling. The main DEGs included ITIH5, GLUL, CACNB2, PDX1, ASPN, PTGER3, MFAP4, PI15, EPHB6, and CPA3 with additional candidates, such as KAZN and the lncRNA gene SSC4D, which have been implicated in migration/invasion, ECM remodeling, or signaling across multiple tumor contexts. Translational analyses in TCGA-BRCA basal-like tumors suggested a descriptive association in which lower LINC01133 levels were accompanied by shifts in the expression trends of genes linked to ECM/EMT programs and modulation of genes related to cell adhesion and protease inhibition. Conclusions: These results suggest a transcriptional model in which LINC01133 is associated with TNBC-related gene expression programs in a concentration-dependent manner, with loss of LINC01133 being associated with a transcriptomic shift toward pro-migratory/ECM remodeling signatures. While functional validation is required to establish causality, these data support LINC01133 as a molecule of interest in breast cancer research. Full article

The tightly controlled and transient acquisition of a motile phenotype by otherwise static epithelial cells (epithelial–mesenchymal transition, EMT) enables the repair of a damaged epithelium. Conversely, a persistent, dysregulated, and exacerbated EMT characterizes epithelial malignancies such as breast carcinoma (BC) and oral squamous cell carcinoma (OSCC), being key for their metastasis and for their escaping anti-tumor immune responses. Herein, we investigated the relationship between EMT signatures and immune cell infiltration across OSCC and metastatic BC with the aim to identify prognostic markers and/or therapeutic targets common to both these malignancies, or unique to OSCC or BC. To this end, we analyzed publicly available transcriptomic datasets to identify coding genes involved in EMT with strong correlation to immune cell signatures. The methodology consisted of data selection, correlation analysis, signature overlap determination, and validation using independent databases. Results indicated that in both OSCC and BC the expression of EMT-related genes is strongly associated with that of immunosuppressive and pro-tumor macrophages. Notably, the FN1 gene coding for the extracellular matrix glycoprotein fibronectin (FN) emerged as the EMT gene common to either tumor types. In confirmation of this, FN protein levels were higher in OSCC and BC tissues than in their normal counterparts. Given FN capability of favoring tumor invasion and metastasis while hindering antitumor immune responses, these data encourage the development of FN antagonists to be used as an adjunct to conventional therapy in the treatment of both OSCC and BC. Full article

Background: Keratoconus (KC) is a progressive corneal ectasia and a leading cause of corneal transplantation in young adults. Once regarded as a biomechanical disorder, KC is now recognized as a complex disease driven by genetic predisposition, epigenetic modulation, and environmental triggers. Advances in genomics and transcriptomics have begun to elucidate the molecular mechanisms underlying corneal thinning and ectasia. Objectives: This review synthesizes two decades of evidence on the genetic and epigenetic architecture of keratoconus, highlights key molecular pathways implicated by these findings, and discusses translational implications for early diagnosis, risk prediction, and novel therapeutic strategies. Methods: A narrative review was conducted of peer-reviewed human, animal, and in vitro studies published from 2000 to 2025, with emphasis on genome-wide association studies (GWAS), sequencing data, methylation profiling, and non-coding RNA analyses. Findings were integrated with functional studies linking genetic variation to molecular and biomechanical phenotypes. Results: Genetic studies consistently implicate loci such as ZNF469, COL5A1, LOX, HGF, FOXO1, and WNT10A, alongside rare variants in Mendelian syndromes (e.g., brittle cornea syndrome, Ehlers–Danlos spectrum). Epigenetic research demonstrates altered DNA methylation, dysregulated microRNAs (e.g., MIR184, miR-143, miR-182), and aberrant lncRNA networks influencing extracellular matrix remodeling, collagen cross-linking, oxidative stress, and inflammatory signaling. Gene–environment interactions, particularly with eye rubbing and atopy, further shape disease expression. Translational progress includes polygenic risk scores, tear-based biomarkers, and early preclinical studies using RNA-based approaches (including siRNA and antisense oligonucleotides targeting matrix-degrading and profibrotic pathways) and proof-of-concept gene-editing strategies demonstrated in corneal cell and ex vivo models. Conclusions: Keratoconus arises from the convergence of inherited genomic risk, epigenetic dysregulation, and environmental stressors. Integrating multi-omic insights into clinical practice holds promise for earlier detection, precision risk stratification, and development of targeted therapies that move beyond biomechanical stabilization to disease modification. Full article

The existing methods used for estimating generator matrixes of BCH codes, which are based on Galois Field Fourier transforms, need to exhaustively test all the possible codeword lengths and corresponding primitive polynomials. With the increase of codeword length, the search space exponentially expands. Consequently, the computational complexity of the estimation scheme becomes very high. To overcome this limitation, a fast estimation method is proposed based on Gaussian elimination. Firstly, the encoded bit stream is reshaped into a matrix according to the assumed codeword length. Then, by using Gaussian elimination, the bit matrix is simplified as the upper triangle form. By testing the independent columns of the upper triangle matrix, the assumed codeword length is judged to be right or not. Simultaneously, by using an augmented matrix, the parity check matrix of a BCH code can be estimated from the simplification result in the procedure of Gaussian elimination. Furthermore, the generator matrix is estimated by using the orthogonality between the generator matrix and parity check matrix. To improve the performance of the proposed method in resisting bit errors, soft-decision data is adopted to evaluate the reliability of received bits, and reliable bits are selected to construct the matrix to be analyzed. Experimental results indicate that the proposed method can recognize BCH codes effectively. The robustness of our method is acceptable for application, and the computation required is much less than the existing methods. Full article

Recent studies have demonstrated that the abundance of brown adipose tissue is inversely associated with obesity in humans. Promoting the browning of white adipocytes therefore represents a promising therapeutic strategy for obesity treatment. LncRNAs are known regulators of adipocyte differentiation and metabolic processes. However, their specific roles in adipocyte browning remain poorly characterized. In this study, we performed transcriptomic analyses using publicly available RNA-seq datasets of mouse white, brown and beige adipose tissues from the EMBL-EBI database. Our analytical workflow included raw data quality control, alignment to the reference genome, transcript assembly, coding potential assessment and differential expression analysis. Functional annotation was conducted through Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses. Key lncRNAs were further validated via Reverse Transcription Quantitative PCR (RT-qPCR). We identified 794 novel lncRNAs and 1499 DEGs, among which 95 were common across all three adipocyte types. Two lncRNAs, MSTRG.12661 and MSTRG.17758, were found to be closely related to critical biological processes, including extracellular matrix organization and fatty acid oxidation. Functional prediction suggests their potential involvement in adipocyte type-specific differentiation. In conclusion, our study reveals novel lncRNAs that may regulate adipocyte differentiation, offering new candidate targets for obesity treatment via induction of white adipose tissue browning. Full article

Background/Objectives: Polycystic ovary syndrome (PCOS) is a common endocrine–metabolic disorder in which skeletal muscle insulin resistance contributes substantially to cardiometabolic risk. Pioglitazone improves insulin sensitivity in women with PCOS, yet the underlying transcriptional changes and their potential as treatment-response biomarkers remain incompletely defined. We aimed to reanalyse skeletal muscle gene expression from pioglitazone-treated PCOS patients using modern machine learning and network approaches to identify candidate biomarkers and regulatory hubs that may support precision therapy. Methods: Public microarray data (GSE8157) from skeletal muscle of obese women with PCOS and healthy controls were reprocessed. Differentially expressed genes (DEGs) were identified and submitted to Ingenuity Pathway Analysis to infer canonical pathways, upstream regulators, and disease functions. Four supervised machine learning algorithms (logistic regression, random forest, support vector machines, and gradient boosting) were trained using multi-step feature selection and 3-fold stratified cross-validation to provide superior Exploratory Gene Analysis. Gene co-expression networks were constructed from the most informative genes to characterize network topology and hub genes. A simulated multi-omics framework combined selected transcripts with representative clinical variables to explore the potential of integrated signatures. Results: We identified 1459 DEGs in PCOS skeletal muscle following pioglitazone, highlighting immune and fibrotic signalling, interferon and epigenetic regulators (including IFNB1 and DNMT3A), and pathways linked to mitochondrial function and extracellular matrix remodelling. Within this dataset, all four machine learning models showed excellent cross-validated discrimination between PCOS and controls, based on a compact gene panel. Random forest feature importance scoring and network centrality consistently prioritized ITK, WT1, BRD1-linked loci and several long non-coding RNAs as key nodes in the co-expression network. Simulated integration of these transcripts with clinical features further stabilized discovery performance, supporting the feasibility of multi-omics biomarker signatures. Conclusions: Reanalysis of skeletal muscle transcriptomes from pioglitazone-treated women with PCOS using integrative machine learning and network methods revealed a focused set of candidate genes and regulatory hubs that robustly separate PCOS from controls in this dataset. These findings generate testable hypotheses about the immunometabolism and epigenetic mechanisms of pioglitazone action and nominate ITK, WT1, BRD1-associated loci and related network genes as promising biomarkers for future validation in larger, independent PCOS cohorts. Full article