Search Results (87)

Background/Objectives: Pacemaker-associated heart failure (PaHF) is a recognized complication of chronic ventricular pacing, yet its long-term incidence and prognostic impact remain incompletely defined. Previous studies on PaHF have been largely limited by small sample sizes, single-center designs, and insufficient long-term or time-dependent analyses. We aimed to evaluate the incidence, clinical predictors, and mortality risk of PaHF in a nationwide real-world cohort. Methods: Using the Korean National Health Insurance Service database, we identified 32,216 patients who underwent de novo pacemaker implantation between 2008 and 2019 without prior heart failure. Results: During a median follow-up of 3.8 years, 4170 patients (12.9%) developed new-onset PaHF and 6184 (19.2%) died. PaHF was independently associated with increased all-cause mortality (adjusted hazard ratio [HR]: 3.11, 95% confidence interval [CI]: 2.93–3.32, p < 0.001), even after accounting for immortal-time bias and relevant covariates. The incidence of PaHF and its associated mortality risk both peaked within the first six months post implantation and remained persistently elevated throughout follow-up; notably, PaHF-associated mortality showed a late resurgence. Sensitivity and subgroup analyses consistently demonstrated higher mortality among patients with PaHF across a wide range of clinical characteristics. Conclusions: In this large, nationwide cohort, the development of PaHF was associated with a substantial and sustained increase in mortality risk following pacemaker implantation. Given the persistent and dynamic nature of this risk, longitudinal monitoring of cardiac function and individualized pacing strategies may be warranted to mitigate long-term adverse outcomes. Additionally, these findings provide real-world benchmarks to guide future pacing strategies and surveillance efforts. Full article

Background: Heart failure (HF) affects over 64 million people globally, imposing substantial morbidity, mortality, and economic burdens. Despite advances in guideline-directed therapies, adherence remains suboptimal due to low health literacy and complex regimens. ChatGPT, an advanced large language model by OpenAI, offers conversational capabilities that could enhance HF education, management, and research. This systematic review synthesizes evidence on ChatGPT’s applications in HF, evaluating its accuracy in patient education and question-answering, enhancing readability, and clinical documentation/symptom extraction. Methods: Following PRISMA guidelines, we searched PubMed, Embase, and Cochrane up to July 2025 using the terms “ChatGPT” and “heart failure”. Inclusion: Studies on ChatGPT (3.5 or 4) in HF contexts, such as in education, readability and symptom extraction. Exclusion: Non-HF or non-ChatGPT AI. Data extraction covered design, objectives, methods, and outcomes. Thematic synthesis was applied. Results: From 59 records, 7 observational studies were included. Themes included patient education/question-answering (n = 5), readability enhancement (n = 2), and clinical documentation/symptom extraction (n = 1). Accuracy ranged 78–98%, with high reproducibility; readability improved to 6th–7th grade levels; and symptom extraction achieved up to 95% F1 score, outperforming traditional machine learning baselines. Conclusions: ChatGPT shows promise in HF care but requires further randomized validation for outcomes and bias mitigation. Full article

Background: Automatic adventitious lung sound classification using deep learning is a promising strategy for objective respiratory disease screening. Evaluating model performance is challenging, particularly with imbalanced clinical datasets. This study compares CNN architectures and proposes a dual-stream classification approach. Methods: Using the public ICBHI 2017 dataset, we compared five pre-trained architectures: VGG16, VGG19, InceptionV3, MobileNetV2, and ResNet152V2. To mitigate class imbalance, we implemented pitch shifting, random shifting, and mixup data augmentation. We also developed and evaluated a novel VGGish-dual-stream network. The primary endpoint was the Average Score (AS), the arithmetic mean of Sensitivity and Specificity. Results: Among benchmarked models, ResNet152V2 achieved the highest AS ($0.541$), approaching the state-of-the-art range (0.56–0.58). This performance was characterised by a high Specificity ($0.67$) but low Sensitivity ($0.41$). Our proposed dual-stream network yielded a more balanced, albeit slightly lower, performance with an AS of $0.508$. Conclusions: Standard CNN architectures like ResNet152V2 can achieve competitive classification performance but may exhibit a clinically significant bias towards high specificity at the expense of sensitivity. This trade-off poses a risk of missing pathological events (false negatives). To ensure clinical safety and utility, future work must prioritise strategies that explicitly improve model sensitivity. Full article

Open-world object detection (OWOD) aims to build detectors that can recognize known categories while simultaneously identifying unknown objects and incrementally learning novel classes. Despite recent advances, existing OWOD approaches still struggle with two critical challenges: the severe bias toward head classes in long-tailed data distributions and the misclassification of unknown objects as background. To address these issues, we introduce TAPM (Tail-Calibrated Transformer Autoencoding with Prototype-Guided Mining), a novel framework that explicitly enhances tail-class representation and robustly reveals unknown objects. TAPM integrates three core innovations: (1) a transformer-based autoencoder that reconstructs region features to calibrate embeddings for rare categories, mitigating the dominance of frequent classes; (2) a prototype-guided mining strategy that leverages class prototypes to localize both overlooked tail instances and candidate unknowns; and (3) an uncertainty-aware soft-labeling mechanism that assigns probabilistic supervision to pseudo-unknowns, reducing noise in incremental learning. Extensive experiments on the MS-COCO and LVIS benchmarks demonstrate that TAPM significantly improves unknown-object recall while maintaining strong known-class accuracy, achieving state-of-the-art performance across both the superclass-separated (S-OWODB) and superclass-mixed (M-OWODB) benchmarks. In particular, TAPM achieves a +20.4-point gain in U-Recall over the strong PROB baseline, underscoring its effectiveness in detecting novel objects without sacrificing mean Average Precision (mAP). Furthermore, TAPM achieves better generalization on cross-dataset evaluations, highlighting its robustness in diverse open-world scenarios. Full article

Background: Healthcare systems globally face complex challenges including rising costs, increasing chronic disease burden, and fragmentation of care. Systems-based models represent promising approaches to healthcare transformation, yet their implementation remains incompletely understood. Objective: To critically analyze the Saudi model of Care (MoC) as a case study of systems-based healthcare transformation, examining its conceptual framework, implementation strategies, and projected health outcomes. Methods: We conducted a narrative review synthesizing publicly available official documents on the Saudi MoC, primarily the 2017 overview and 2025 revision, identified through targeted searches of Ministry of Health websites and grey literature portals (no date restrictions); formal quality appraisal was not applied as sources were official policy documents, with bias mitigated through cross-verification and critical analysis. Results: The Saudi MoC exemplifies systems-based transformation through its multi-layered framework organized around six patient-centered systems of care spanning the lifecycle. Key innovations include: (1) an architectural approach integrating activated individuals, healthy communities, virtual care, and traditional clinical settings; (2) a comprehensive intervention taxonomy with 42 specific initiatives; (3) explicit contextual adaptations for diverse settings; and (4) a phased implementation approach with detailed performance metrics. National indicators improved during the reform period, including life expectancy and maternal and child health. These are national trends observed during the period of health reforms. Causal attribution to the Model of Care requires a counterfactual evaluation. Conclusions: This analysis of the Saudi MoC contributes to the literature on systems-based healthcare transformation by illuminating how theoretical principles can be operationalized at national scale. The model’s patient-centered design, comprehensive intervention taxonomy, and attention to implementation factors offer valuable insights for other healthcare systems pursuing transformation. Further research should examine actual implementation outcomes as the model matures. Full article

Background/Objectives: Multiple sclerosis (MS) is a chronic demyelinating disease where early identification of patients at risk of conversion from clinically isolated syndrome (CIS) to clinically definite MS remains a critical unmet clinical need. Existing machine learning approaches often lack interpretability, limiting clinical trust and adoption. The objective of this research was to develop a novel two-stage machine learning framework with comprehensive explainability to predict CIS-to-MS conversion while addressing demographic bias and interpretability limitations. Methods: A cohort of 177 CIS patients from the National Institute of Neurology and Neurosurgery in Mexico City was analyzed using SeruNet-MS, a two-stage framework that separates demographic baseline risk from clinical risk modification. Stage 1 applied logistic regression to demographic features, while Stage 2 incorporated 25 clinical and symptom features, including MRI lesions, cerebrospinal fluid biomarkers, electrophysiological tests, and symptom characteristics. Patient-level interpretability was achieved through SHAP (SHapley Additive exPlanations) analysis, providing transparent attribution of each factor’s contribution to risk assessment. Results: The two-stage model achieved a ROC-AUC of 0.909, accuracy of 0.806, precision of 0.842, and recall of 0.800, outperforming baseline machine learning methods. Cross-validation confirmed stable performance (0.838 ± 0.095 AUC) with appropriate generalization. SHAP analysis identified periventricular lesions, oligoclonal bands, and symptom complexity as the strongest predictors, with clinical examples illustrating transparent patient-specific risk communication. Conclusions: The two-stage approach effectively mitigates demographic bias by separating non-modifiable factors from actionable clinical findings. SHAP explanations provide clinicians with clear, individualized insights into prediction drivers, enhancing trust and supporting decision making. This framework demonstrates that high predictive performance can be achieved without sacrificing interpretability, representing a significant step forward for explainable AI in MS risk stratification and real-world clinical adoption. Full article

Background/Objectives: The voltage-gated potassium channels of the Kv4 family (Kv4.1, Kv4.2, Kv4.3) regulate neuronal excitability and synaptic integration. The dysregulation of these channels has been linked to neurodegenerative diseases, such as Alzheimer’s disease (AD), spinocerebellar ataxias, amyotrophic lateral sclerosis (ALS), prion diseases, and Parkinson’s disease (PD). Current evidence is scattered across diverse models, and a systematic synthesis is lacking. This review seeks to compile and analyze data on Kv4 channel alterations in neurodegeneration, focusing on genetic variants, functional changes, and phenotypic consequences. Methods: A systematic search was conducted for peer-reviewed studies, including human participants, human-derived cell models, and relevant animal models. Studies were considered eligible if they investigated Kv4.1–Kv4.3 (encoded by gene encoding the Kv4.1-Kv4.3 α-subunit of voltage-gated A-type potassium channels (KCND1-KCND3)) expression, function, or genetic variants, as well as associated auxiliary subunits such as DPP6 (dipeptidyl peptidase–like protein 6) and KChIP2 (Kv channel–interacting protein 2), in neurodegenerative diseases. Both observational and experimental designs were considered. Data extraction included disease type, model, Kv4 subunit, functional or genetic findings, and key outcomes. Risk of bias was assessed in all included studies. Results: Kv4 channels exhibit significant functional and expression changes in various neurodegenerative diseases. In AD and prionopathies, reduced Kv4.1- and Kv4.2-mediated currents contribute to neuronal hyperexcitability. In spinocerebellar ataxias, KCND3 mutations cause loss- or gain-of-function phenotypes in Kv4.3, disrupting cerebellar signaling. In models of ALS and PD, Kv4 dysfunction correlates with altered neuronal excitability and can be modulated pharmacologically. Subunit modulators such as DPP6 and KChIP2 influence channel function and could represent therapeutic targets. Conclusions: Kv4 channels are crucial for neuronal excitability in multiple neurodegenerative contexts. Dysregulation through genetic or pathological mechanisms contributes to functional deficits, highlighting Kv4 channels as promising targets for interventions aimed at restoring electrical homeostasis and mitigating early neuronal dysfunction. Full article

Background/Objectives: Triple-negative breast cancer (TNBC) is an aggressive subtype lacking ER, PR, and HER2 expression, with limited targeted therapies and poor outcomes. Epigenetic dysregulation, particularly aberrant DNA methylation, is a key driver. Decitabine, a DNA methyltransferase inhibitor (DNMTi), shows promise by reactivating silenced tumor suppressor genes and modulating immune responses. This systematic review evaluates preclinical and clinical evidence on decitabine’s efficacy, mechanisms, and translational potential in TNBC. Methods: A PRISMA-2020 compliant search of PubMed, EBSCO, Web of Science, and Semantic Scholar was conducted up to April 2025. Included studies assessed decitabine alone or in combination in TNBC preclinical or clinical settings. Risk of bias was assessed using QUIPS and RoB 2.0 tools. Results: Twenty-five studies were included. In vitro, decitabine-induced growth inhibition, apoptosis, and re-expression of silenced genes (such as BRCA1 and CDH1). In vivo, it reduced tumor burden and enhanced anti-tumor immunity through MHC-I, PD-L1, and STING pathway upregulation. Synergy was noted with anti-PD-1, HDAC inhibitors, and chemotherapy. Resistance mechanisms included persistent DNMT activity, low DCK, and miRNA-driven escape (miR-155–TSPAN5). Conclusions: Decitabine demonstrates strong preclinical and early clinical potential in TNBC via epigenetic reprogramming and immune activation. Future strategies should focus on biomarker-based selection and resistance mitigation. Full article

With the expansion of utility-scale renewable energy development worldwide, accurate estimation of bird and bat fatalities is needed for informed policy-making and appropriate formulation of mitigation strategies. Background mortality, or the mortality caused by natural as opposed to anthropogenic processes, is often identified as a positive bias, and sometimes it is identified as a substantial or even leading contributor to fatality estimates. To estimate background mortality, I compiled fatalities/ha counted during searches of turbine-free study sites reported by others over 2548 ha and myself over 2297 ha. No bat fatalities were found in any of these searches. Bird fatalities/ha averaged 0.0055. I also compared estimates of fatalities/ha before and after turbine removals from 123 rows of wind turbines in California’s Altamont Pass Wind Resource Area (APWRA). These turbine rows had been searched for fatalities over various periods during 1998–2002 and 2006–2014, and fatalities had been recorded at each row during first searches of new monitoring periods. I used the same search methods as the monitor, but my first searches covered 624 ha of plots centered around vacant turbine sites. I found 0.0194 (95% CI: 0.0035–0.0352) bird fatalities/ha, but no bat fatalities. I estimated that background mortality was 3.6% (95% CI: 0–6.2%), mortality caused by unremoved power lines and meteorological towers was 8.2% (95% CI: 0–15.8%), and mortality caused by wind turbines was 88.2% (95% CI: 78–100%). Contamination of carcasses from operable wind turbines ≥ 400 m distant from vacant turbine sites likely biased my estimate upward by 3.5-fold compared to natural mortality averaged among sites far from wind turbines. This study does not support the notion that background mortality contributes substantially to mortality estimates at renewable energy projects. Full article

Background: This systematic review examines the lived experiences of Black students in UK higher education (HE), focusing on their encounters with racism and racial disadvantage, and how institutional and social factors contribute to these experiences. Methods: We conducted a systematic search across seven databases (Academic Search Complete, Education Abstracts, PsycINFO, Race Relations Abstracts, Scopus, Web of Science, and SocINDEX) in April 2023, with periodic updates. The grey literature, which refers to research and information produced outside of traditional academic publishing and distribution channels, was reviewed. This includes reports, policy briefs, theses, conference proceedings, government documents, and materials from organisations, think tanks, or professional bodies that are not commercially published or peer-reviewed but can still offer valuable insights relevant to the topic. Hand searches were also included. Studies were included if they were peer-reviewed, published between 2012 and 2024, written in English, and focused on the experiences of Black students in UK higher education. Both qualitative and quantitative studies with a clear research design were eligible. Studies were excluded if they lacked methodological rigour, did not focus on the UK HE context, or did not disaggregate Black student experiences. Risk of bias was assessed using standard qualitative appraisal tools. Thematic analysis was used to synthesise findings. Results: Nineteen studies were included in the review. Two main themes emerged: (1) diverse challenges including academic barriers and difficulties with social integration, and (2) the impact of racism and institutional factors, such as microaggressions and biased assessments. These issues contributed to mental fatigue and reduced academic performance. Support systems and a sense of belonging helped mitigate some of the negative effects. Discussion: The evidence was limited by potential bias in reporting and variability in study quality. Findings reveal persistent racial inequalities in UK HE that affect Black students’ well-being and outcomes. Institutional reforms, increased representation, and equity-focused policies are needed. Future research should explore effective interventions to reduce the awarding gap and support Black student success Full article

Background: While insulin is the standard of care for feline diabetes mellitus (FDM), non-insulin antidiabetic drugs (NIADs) are emerging as alternatives. This systematic review aims to synthesize and critically appraise the current evidence for the efficacy and safety of NIADs in cats. Methods: A systematic review was conducted following PRISMA guidelines. Major databases were searched for studies evaluating NIADs in diabetic cats or at-risk/experimental models. Risk of bias was assessed using RoB 2 and ROBINS-I tools. Results: Twenty studies were included. In diabetic cats (10 studies), traditional agents (glipizide, metformin, acarbose) showed limited efficacy based on evidence with a high risk of bias. Newer SGLT2 inhibitors (bexagliflozin, velagliflozin) demonstrated high treatment success rates and non-inferiority to insulin but were associated with a significant risk of euglycemic diabetic ketoacidosis (eDKA). In at-risk/experimental models (10 studies), thiazolidinediones consistently improved insulin sensitivity, while glipizide was shown to accelerate islet amyloidosis. Conclusions: The evidence supports a paradigm shift towards SGLT2 inhibitors as a viable oral monotherapy for select cases of FDM. This alters the clinical risk–benefit discussion from preventing hypoglycemia to mitigating eDKA. Significant evidence gaps remain, particularly the lack of high-quality RCTs and data in cats with common comorbidities. Full article

Background: Social determinants of health (SDOH) are critical contributors to cancer disparities, influencing prevention, early detection, treatment access, and survival outcomes. Addressing these disparities is essential in achieving equitable oncology care. Artificial intelligence (AI) is revolutionizing oncology by leveraging advanced computational methods to address SDOH-driven disparities through predictive analytics, data integration, and precision medicine. Methods: This review synthesizes findings from systematic reviews and original research on AI applications in cancer-focused SDOH research. Key methodologies include machine learning (ML), natural language processing (NLP), deep learning-based medical imaging, and explainable AI (XAI). Special emphasis is placed on AI’s ability to analyze large-scale oncology datasets, including electronic health records (EHRs), geographic information systems (GIS), and real-world clinical trial data, to enhance cancer risk stratification, optimize screening programs, and improve resource allocation. Results: AI has demonstrated significant advancements in cancer diagnostics, treatment planning, and survival prediction by integrating SDOH data. AI-driven radiomics and histopathology have enhanced early detection, particularly in underserved populations. Predictive modeling has improved personalized oncology care, enabling stratification based on socioeconomic and environmental factors. However, challenges remain, including AI bias in screening, trial underrepresentation, and treatment recommendation disparities. Conclusions: AI holds substantial potential to reduce cancer disparities by integrating SDOH into risk prediction, screening, and treatment personalization. Ethical deployment, bias mitigation, and robust regulatory frameworks are essential in ensuring fairness in AI-driven oncology. Integrating AI into precision oncology and public health strategies can bridge cancer care gaps, enhance early detection, and improve treatment outcomes for vulnerable populations. Full article

Vision–Language Models (VLMs) have become key contributors to the state of the art in contextual emotion recognition, demonstrating a superior ability to understand the relationship between context, facial expressions, and interactions in images compared to traditional approaches. However, their reliance on contextual cues can introduce unintended biases, especially when the background does not align with the individual’s true emotional state. This raises concerns for the reliability of such models in real-world applications, where robustness and fairness are critical. In this work, we explore the limitations of current VLMs in emotionally ambiguous scenarios and propose a method to overcome contextual bias. Existing VLM-based captioning solutions tend to overweight background and contextual information when determining emotion, often at the expense of the individual’s actual expression. To study this phenomenon, we created synthetic datasets by automatically extracting people from the original images using YOLOv8 and placing them on randomly selected backgrounds from the Landscape Pictures dataset. This allowed us to reduce the correlation between emotional expression and background context while preserving body pose. Through discriminative analysis of VLM behavior on images with both correct and mismatched backgrounds, we find that in 93% of the cases, the predicted emotions vary based on the background—even when models are explicitly instructed to focus on the person. To address this, we propose a multimodal approach (named BECKI) that incorporates body pose, full image context, and a novel description stream focused exclusively on identifying the emotional discrepancy between the individual and the background. Our primary contribution is not just in identifying the weaknesses of existing VLMs, but in proposing a more robust and context-resilient solution. Our method achieves up to 96% accuracy, highlighting its effectiveness in mitigating contextual bias. Full article

Background: This systematic review explores the integration of green practices into humanitarian supply chains to mitigate environmental impacts and contribute to global decarbonization efforts. Methods: This review focused on peer-reviewed articles published between 2011 and 2024 that addressed the environmental dimension of humanitarian logistics. Studies were included if they examined environmental practices within humanitarian supply chains and excluded if they lacked focus on environmental impact or logistics. A comprehensive search of the Scopus database in April 2024 yielded 291 records, of which 51 studies met the inclusion criteria. A thematic synthesis was conducted; due to the qualitative nature of the data, no formal risk-of-bias assessment was conducted. Results: The analysis revealed increasing adoption of environmentally focused practices, such as emissions monitoring, waste reduction, and resource-efficient transportation. Key barriers included operational complexity, inadequate digital infrastructure, and the absence of standardized environmental frameworks. The review identified digital innovation, inter-organizational collaboration, and integrated environmental performance metrics as promising pathways for improvement. Despite growing awareness, significant gaps remain in the standardization and measurement of environmental performance across humanitarian supply chains. Conclusions: The findings highlight the need for further research and coordinated efforts to develop consistent, scalable green practices in the humanitarian context. Full article

Background: Aluminum accumulation in the brain causes cognitive deficits. No comprehensive synthesis of pharmacological treatments against aluminum neurotoxicity has been conducted, which led us to systematically review the effects of various pharmacological agents against aluminum-induced neurotoxicity, primarily addressing learning and memory after chronic aluminum exposure (≥2 months) in rodent models. Methods: A literature search was performed in PubMed, Google Scholar, Science Direct, and Scopus for studies published between 2000 and 2023. A total of 45 studies were selected according to the inclusion criteria. Primary outcomes focused on assessing learning and memory, with 39 different pharmacological agents evaluated explicitly for their effects against aluminum-induced neurotoxicity. Meta-analysis and subgroup analysis were performed to evaluate cognitive improvement in the Morris water maze (MWM) for learning and memory, and oxidative stress parameters were evaluated through superoxide dismutase (SOD) and catalase (CAT) in aluminum-induced neurotoxicity models. Results: According to the systematic analysis, most treatments significantly improve learning and memory, except for insulin and melatonin. According to the MWM analysis, Memantine, Hypericum perforatum extract, Bennincasa hespidia, and, based on the biochemical analysis, Chrysin showed better results. The meta-analysis (random effects) revealed reduced escape latency (SMD = 0.97, 95% CI: 0.74 to 1.19) and increased SOD (SMD = −0.54, 95% CI: −0.79 to −0.29) and CAT levels (SMD = −0.50, 95% CI: −0.73 to −0.27) in treated groups versus aluminum. Egger’s regression tests showed no strong evidence of publication bias. Conclusions: This study effectively synthesized preclinical evidence, identifying promising pharmacological agents for mitigating aluminum-induced cognitive deficits. These findings offer a scientific basis for future experimental studies and therapeutic development targeting aluminum neurotoxicity. Full article