SEARCH · Search PubMed

Results for “Predictive Value of Tests”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

1,066 records · Page 2Linked to original sources

PubMed2026

Machine learning-based prediction of unplanned readmission and construction of an online calculator for elderly patients with mild ischemic stroke.

OBJECTIVE: To screen for independent risk factors for unplanned readmission in elderly patients with mild ischemic stroke, and to construct and validate an online risk prediction calculator based on an interpretable machine learning model, thereby providing a promising practical tool for accurate clinical assessment of 30‑day all‑cause unplanned readmission risk in this population. METHODS: A prospective cohort study was conducted, including 1050 patients aged ≥ 60 years with mild ischemic stroke admitted between August 2023 and September 2024. Participants were randomly divided into a training set (840 cases) and a test set (210 cases) at a ratio of 8:2. Risk factors were screened by univariate analysis and multivariable Logistic regression. Four machine learning models, namely LightGBM, XGBoost, Random Forest, and K‑Nearest Neighbors (KNN), were developed and their performance was evaluated using AUC, accuracy, sensitivity, and specificity as metrics. The SHAP framework was used for interpretability analysis, and an online calculator was subsequently developed based on the optimal model. RESULTS: Univariate analysis showed significant differences (P < 0.05) in 13 factors including age, smoking, AIP, TyG index, HALP score, etc. Multivariable Logistic regression identified age (OR = 9.752), smoking (OR = 5.171), AIP (OR = 6.691), TyG index (OR = 4.393), HALP score (OR = 2.831), and ≥ 2 comorbidities (OR = 3.664) as independent risk factors. All four machine learning models demonstrated good predictive performance. Based on a comprehensive evaluation of multiple metrics and computational efficiency, the LightGBM model exhibited the best predictive performance (AUC = 0.884, accuracy = 0.829, sensitivity = 0.812, specificity = 0.875). SHAP analysis showed that age, AIP, TyG index, smoking, and HALP score were key predictors. An online calculator developed based on this model enables individualized risk predictions. CONCLUSION: Key risk factors associated with 30‑day unplanned readmission in elderly patients with mild ischemic stroke were identified. The LightGBM model demonstrated high predictive accuracy, and together with the interpretability analysis and online calculator, offers a practical tool to support clinical risk assessment. However, this tool requires future external validation.

Humans↗

PubMed2025

Application of causal discovery of factors driving dissolved oxygen in estuarine environments.

Dissolved oxygen (DO) concentrations in estuarine bottom waters are a manifestation of multiple, interacting physical and biogeochemical processes, yet identifying their independent contributions remains challenging. Here, we analyze monthly water quality monitoring data from eight stations across Long Island Sound from 1994 to 2022 using a causal discovery framework (PCMCI+) and transformation of forcing variables. Our goal is to identify and isolate variables that causally influence bottom DO and improve predictive models by minimizing overfitting and multicollinearity. PCMCI+ reveals surface-layer temperature as the most important and consistent negative driver of bottom DO, followed by stratification. Wind events exhibit only brief relief by advection and mixing, while river discharge shows no direct causal link to DO, making it less influential than previously thought. Biogeochemical variables, including chlorophyll-a (Chl-a), nitrate and nitrite, and particulate carbon, influence DO through both contemporaneous and time-lagged pathways, often with signs that shift depending on the process. The derived models were evaluated by comparing skill scores, mean squared error, and Akaike Information Criterion. Both model types perform well, with coefficient of determination values exceeding 0.90 at multiple stations using only 3-5 predictors. Our analysis reveals that the best causal predictors are surface-layer temperature, stratification, Chl-a, and particle carbon. This approach provides a scalable framework for improving prediction models and understanding the mechanistic links that control the seasonal variability of DO in estuarine systems.

Estuaries↗

PubMed2026

Non-parametric differential methylation analysis characterizes histotype-specific promoter regions in epithelial ovarian cancer.

Epithelial ovarian cancer (EOC) is a heterogenous disease with frequent late-stage diagnosis and high mortality rates, for which no reliable screening tests exist. In recent years, epigenetic biomarkers in the form of DNA methylation in CpG-rich regions have gained increased attention in the scientific community due to their robust nature and accessibility, allowing for diagnosis without the need for invasive surgery. In this study, we investigated the aberrant methylation of promoter regions in early stage EOC through non-parametric methods, with the purpose of characterizing candidate epigenetic biomarkers. The approach was used on a cohort of early stage EOC samples, and results were compared to existing programs for differential methylation. Significant regions were then used to construct a CpG panel for stratifying EOC histotypes through predictive classification in external data. Identified promoter regions were highly reproducible across cohorts, and the constructed CpG model stratified histotypes in external cohorts through predictive classification. Comparisons against other DMP and DMR callers showed a degree of homogeneity between results but also revealed promoter regions that were overlooked despite clear signs of aberrant methylation. Finally, EOC histotypes were found to differ in their methylation distribution types, and results indicate that methods sensitive to non-normally distributed data may be poorly suited to compare groups with different distribution types. The non-parametric approach identified aberrantly methylated promoter regions that were highly reproducible across cohorts. Results from predictive classification indicate that these regions may be useful for the purpose of EOC histotype stratification.

Humans↗

PubMed2026

Does impulsivity predict treatment outcomes in PTSD with borderline personality disorder features? Results from a randomized clinical trial.

BACKGROUND: Trauma-focused psychotherapies are first-line treatments for posttraumatic stress disorder (PTSD). However, a substantial proportion of clients do not respond adequately or drop out of therapy prematurely. This has sparked interest in identifying individual-level predictors of treatment outcomes, including improvement in PTSD severity and dropout. Impulsivity may be a predictor because it may interfere with key therapeutic processes, such as cognitive restructuring and emotional processing. Consequently, we present a hypothesis-driven secondary analysis of a 15-month randomized clinical trial comparing Dialectical Behavior Therapy for PTSD (DBT-PTSD) and Cognitive Processing Therapy (CPT) in women with childhood abuse-related PTSD and borderline personality disorder features to test whether impulsivity, assessed at baseline, predicts PTSD improvement and dropout. We further explore whether the dimensions of impulsivity (non-planning, attentional impulsivity, and motor impulsivity) differentially affect the outcomes in DBT-PTSD vs. CPT. METHODS: A total of 193 cis women with PTSD related to childhood abuse and borderline personality disorder features were assessed using the Clinician-Administered PTSD Scale (CAPS) and the Barratt Impulsiveness Scale (BIS-10). Separate probit models and general linear models were applied to predict dropout and pre-to-post changes in PTSD severity (ΔCAPS) from total impulsivity and subscale scores, i.e. non-planning, attentional and motor impulsivity. RESULTS: Overall, dropout rates were higher for participants with higher baseline impulsivity scores (p = 0.049), particularly for those with higher non-planning impulsivity (p = 0.012). In participants randomized to CPT improvement in PTSD symptom severity (ΔCAPS) was negatively related to baseline total impulsivity (p = 0.021). In participants randomized to DBT-PTSD this relation was not significant. CONCLUSIONS: The results suggest that impulsivity may predict treatment outcomes. Specifically, patients with elevated impulsivity may be less likely to respond adequately to CPT. If replicated, these findings have implications for personalization of treatment.

Humans↗

PubMed2026

Quo vadis, BGA? A collaborative EDNAP exercise on the challenges and progress in forensic biogeographical ancestry inference.

There is a broad consensus that forensic tests for the prediction of externally visible characteristics (EVC) and analysis of biogeographic ancestry (BGA) of an individual are technically reliable. However, interpretation of the results and population-specific genotype distribution patterns remains challenging. EVC and BGA analyses provide valuable information for population genetics studies and as investigative leads for criminal cases, as well as for historical and contemporary identification tests. However, inaccurate or incorrect predictions, for example, from subjective bias in the interpretations made, have the potential to misdirect police investigations. The legal situation regarding EVC and BGA testing varies by country: ranging from countries where it is explicitly prohibited, to those without specific regulations on biogeographic ancestry prediction, and others that have already enacted laws governing its use. The reluctance to utilize these analyses is not only due to legal restrictions and data protection concerns, but also to initial limited sets of sufficiently comprehensive forensic DNA assays. Forensic BGA marker panels typically contain up to ∼300 SNPs. This relatively small number of genetic markers, along with limited reference population data, complicates the interpretation of results from donors of unknown origin. This paper presents the results of a collaborative EDNAP study, which, for the first time, evaluated the approach to reporting EVC and BGA data between international laboratories. For the study, DNA from nine individuals with self-reported ancestry was collected and analysed using various forensic panels differing in the number and composition of ancestry-informative markers genotyped, comprising: the Precision ID mtDNA Whole Genome Panel, the VISAGE Basic Tool and the VISAGE Enhanced Tool for Appearance and Ancestry Prediction, and the Ion AmpliSeq™ PhenoTrivium Panel. To ensure full data protection, all SNP genotypes and uniparental marker haplotypes obtained were not shared with third parties. Instead, the genetic data were analysed using a range of commonly used population analysis software packages. These analysis outcomes were then distributed to twelve European forensic laboratories (both academic and law enforcement institutions), who were asked to prepare reports based on their interpretation of the phenotypes and ancestry they inferred from the analysis data. A questionnaire sent alongside the genetic information, aimed to evaluate which difficulties were encountered by the participants in processing the BGA analysis data they were given.

Humans↗

PubMed2026

Diagnostic performance of machine learning models for malignant and non-malignant pleural effusion: Systematic review and meta-analysis.

BACKGROUND: Accurately distinguishing malignant pleural effusion (MPE) from non-malignant pleural effusion is clinically important, but the generalisability and methodological quality of machine-learning (ML) models remain uncertain. METHODS: We searched eight databases to 23 April 2026. Diagnostic performance was pooled using random-effects and Reitsma bivariate models, and study quality was assessed using PROBAST+AI. RESULTS: Forty-two studies were included; 17 contributed to the AUC meta-analysis and 14 to the bivariate analysis. The pooled AUC was 0.90 (95 % CI 0.85-0.94; 95 % prediction interval 0.62-0.98), with sensitivity of 0.80 (95 % CI 0.77-0.83) and specificity of 0.87 (95 % CI 0.79-0.92). Only nine studies reported external, temporal or independent validation. Externally validated studies had a lower pooled AUC than studies without external validation (0.83 vs 0.92), with lower specificity observed in the two externally validated studies contributing sensitivity and specificity data. All 42 development assessments had high overall quality concerns, and all 42 model evaluations were judged at high risk of bias. CONCLUSIONS: ML models showed good apparent accuracy for distinguishing MPE from non-MPE, but the evidence was limited by substantial heterogeneity, high risk of bias and scarce external validation. The pooled estimates reflect the average performance of different selected models rather than the expected accuracy of a single clinical test. ML models should be regarded as adjuncts to existing diagnostic pathways until they are confirmed by rigorous multicentre prospective external validation and clinical-impact studies.

Humans↗

PubMed2026

Multi-omic biomarkers in cardiovascular disease: Discovery to clinical translation.

Cardiovascular disease (CVD) remains the leading cause of mortality worldwide, necessitating improved risk stratification and early detection strategies. Multiomics approaches that integrate genomics, transcriptomics, proteomics, metabolomics, and epigenomics offer unprecedented opportunities for biomarker discovery and precision medicine in cardiovascular care. This narrative review examines the current landscape of multiomics biomarkers for CVD, tracing their evolution from discovery to clinical translation. We synthesize evidence from recent studies evaluating the clinical utility of integrated omics approaches across diverse cardiovascular conditions, including atherosclerotic cardiovascular disease, heart failure, and atrial fibrillation. High-throughput proteomics has identified novel protein signatures that enhance cardiovascular risk prediction beyond traditional risk factors. Metabolomics has revealed pathway-specific biomarkers, including trimethylamine N-oxide and lipid species, associated with atherogenesis. Polygenic risk scores derived from genomic data demonstrate incremental value when combined with clinical risk scores. Multiomics biomarkers represent a transformative approach to cardiovascular risk assessment and disease management.

Humans↗

PubMed2026

Mobile health apps improve Health-Related Quality of Life in Type 2 Diabetes Mellitus by enhancing medication adherence: A multicentre randomised controlled trial with mediation analysis.

AIMS: This study evaluated whether a gamified mHealth application (CareAide®) improves Health-Related Quality of Life (HRQoL) in Type 2 Diabetes Mellitus (T2DM) and whether this effect is mediated by medication adherence. METHODS: Prespecified secondary analysis of the T2DM cohort from a 6-month multicentre RCT (NCT06068309; N = 663; three Malaysian hospitals). Participants were randomised 1:1 to standard care or CareAide®. Adherence (MMAS-8), EQ-5D-5L utility (Malaysian value set), and AQoL-6D were assessed at baseline and 6 months. Simple mediation analysis (PROCESS Model 4; 5000 bootstraps) adjusted for baseline HRQoL. RESULTS: CareAide® significantly predicted higher MMAS-8 scores (mean difference +1.756; d = 1.638; p < 0.001). Higher MMAS-8 scores significantly predicted improved AQoL-6D utility (b = 0.024; p < 0.001). The direct effect on AQoL-6D was non-significant (p = 0.248). Bootstrapped indirect effect confirmed full mediation via AQoL-6D (0.042; 95% CI [0.024, 0.060]). A sensitivity analysis adjusting for baseline HbA1c confirmed full mediation (indirect = 0.034; 95% CI [0.015, 0.052]; n = 563). EQ-5D-5L utility showed a significant direct between-group difference at 6 months (p = 0.012) but did not operate as a mediation outcome. CONCLUSIONS: Medication adherence fully mediates the AQoL-6D HRQoL benefit of a gamified mHealth intervention in T2DM, as confirmed by both the primary and HbA1c-adjusted sensitivity analyses. These findings support integration of behaviourally informed digital adjuncts into routine primary diabetes care.

Humans↗

PubMed2026

Recent advances in Strongyloides screening, diagnostics, therapeutics, and management.

PURPOSE OF REVIEW: Strongyloidiasis affects an estimated 30-100 million people globally and can have life-threatening consequences in immunocompromised hosts, yet it remains underdiagnosed due to limited access and performance of available diagnostics. Novel assays and anthelmintics may reshape screening, diagnosis, treatment, and prevention for at-risk populations. RECENT FINDINGS: Advances in molecular diagnostics coupled with robust stool extraction methods have supplanted traditional parasitologic methods in settings where nucleic acid amplification is feasible. Transition from standard immunoglobulin G (IgG)-based immunoassays to the new IgG- and IgG4-based rapid diagnostic tests using recombinant Strongyloides stercoralis nematode immunodominant E antigen (NIE) and/or S. stercoralis immunoreactive antigen (SsIR) has facilitated serologic screening at the point of care. The World Health Organization now conditionally recommends community-wide ivermectin mass drug administration in highly endemic settings. Regarding new treatment options, moxidectin is noninferior to ivermectin with 93-94% cure rates and a longer half-life, while emodepside shows 80-90% predicted cure rates in early trials and offers a mechanistically distinct option. Understanding of immunosuppressed populations at risk for hyperinfection has expanded, prompting updated screening recommendations. SUMMARY: Serologic and molecular tools are improving screening and diagnosis, and moxidectin and emodepside may broaden treatment options, but data in severe disease and special populations remain limited. Priorities include harmonized screening algorithms and prospective studies in high-risk groups.

Humans↗

PubMed2026

Efferocytosis regulatory factors in atherosclerosis: A preclinical systematic review.

BACKGROUND: Impaired efferocytosis is a key driver of plaque instability during atherosclerosis progression. Efficient clearance of apoptotic cells through efferocytosis relies on the coordinated action of multiple regulatory factors. METHODS: PubMed, Web of Science, ScienceDirect, OVID MEDLINE, and Scopus were searched for studies published up to February 7, 2026. Eligible preclinical studies were systematically reviewed to identify endogenous factors that regulate efferocytosis in atherosclerosis. Clinical evidence was also incorporated to enable a preliminary translational assessment of these regulatory factors. RESULTS: Thirty-five endogenous regulatory factors were identified from 36 included studies, and their functional roles across distinct stages of efferocytosis were characterized. Notably, metabolic regulators such as PKM2, PFKFB3, GLS1, and Drp1 were involved in distinct efferocytosis stages. This suggests that metabolic reprogramming may provide the metabolic support require for efficient efferocytosis and inflammation resolution. Ten factors were supported by preliminary clinical evidence consistent with preclinical data. PKM2 was the only candidate biomarker with prospective observational data. However, its independent predictive value still requires validation in multicenter prospective studies. CONCLUSIONS: This review provides a systematic synthesis of 35 endogenous efferocytosis regulators and elucidates their regulatory network in atherosclerosis based on a functional stage framework. Metabolic reprogramming is identified as a central hub linking efferocytosis efficiency to inflammation resolution. This review offers a new theoretical basis for efferocytosis-targeted intervention strategies.

Animals↗

PubMed2026

The psychosocial supports and interventions accessed by family members of patients with developmental and epileptic encephalopathies: A systematic review.

AIM: Family members of patients with developmental and epileptic encephalopathies (DEEs) face profound emotional, social and practical challenges, yet little is known about how they access psychosocial support. We synthesised the literature on the psychosocial support accessed by family members of patients with DEEs, including evidence-based interventions. METHOD: Four databases were searched. Two reviewers independently screened and extracted data, appraised study quality (QualSyst Tool), and determined certainty of evidence (GRADE-CERQual Framework). Data were synthesised using inductive thematic analysis. RESULTS: 28 papers comprising 27 unique studies were included and methodological quality was high overall (median = 0.91, IQR = 0.85-1.00). Most studies focused on parent experiences; only two included siblings' perspectives. Families reported high psychological needs which were rarely met by psychological support. Clinicians rarely provided adequate information, quality communication or addressed mental health. Families valued respite care, but experienced barriers to access. Peer support was the most common and valued resource. Three studies tested interventions and appeared feasible and acceptable for parents. However, evidence for intervention effectiveness is preliminary and limited to small, uncontrolled pilot studies. INTERPRETATION: Gaps exist in psychosocial support provision for families of patients with DEEs, including insufficient research on siblings' and grandparents' needs, and lack of evidence-based interventions. We propose evidence-informed research and implementation strategies to address these gaps.

Humans↗

PubMed2026

Triazole resistance in clinical Aspergillus fumigatus isolates in India, a multicenter surveillance study.

BACKGROUND: Triazole resistance in Aspergillus fumigatus is a global public health concern associated with treatment failure, notably in invasive aspergillosis. However, population-level data on triazole resistance from India remain limited, with most reports originating from single-center studies. METHODS: We conducted a multicenter surveillance study to assess the prevalence of triazole resistance among clinical A. fumigatus isolates across India. Antifungal susceptibility testing was performed using the CLSI broth microdilution method (M38-Ed3), and molecular characterization was conducted on resistant isolates. A total of 518 isolates were analyzed: 115 prospectively collected from 13 tertiary-care hospitals from 2015-2020, and 403 archived isolates obtained from the National Culture Collection of Pathogenic Fungi (1994-2020). RESULTS: The overall pooled prevalence of non-wildtype isolates was 4.1% for itraconazole (95% CI: 2.54-6.17%), 3.9% for posaconazole (95% CI: 2.39-5.94%), while 1.4% were resistant to voriconazole (95% CI: 0.55-2.77%). One multi-azole-resistant isolate from an immunocompromised, mold-active triazole-naïve patient carried the TR34/L98H mutation, suggesting environmental acquisition. Prevalence of resistance did not differ significantly across geographic regions or between public and private sector hospitals. Linear regression analysis revealed a significant temporal increase in median MICs of all three licensed triazoles between 1994 and 2020. Approximately 29% of isolates exhibited amphotericin B MICs exceeding the epidemiological cutoff value; however, the clinical significance of this finding remains uncertain. CONCLUSIONS: Azole resistance among clinical A. fumigatus isolates in India remains uncommon (<5%), supporting the continued use of triazoles as first-line therapy. However, the observed temporal increase in triazole MICs underscores the need for sustained national surveillance to detect emerging resistance trends.

Aspergillus fumigatus↗

PubMed2026

PPRC1 is a prognostic biomarker and key regulator of mitochondrial oxidative phosphorylation in multiple myeloma.

BACKGROUND: Multiple myeloma (MM) remains an incurable haematological malignancy, underscoring the need for novel prognostic biomarkers and therapeutic targets. This study aimed to investigate the clinical and biological significance of peroxisome proliferator-activated receptor gamma coactivator-related protein 1 (PPRC1) in MM. METHODS: Expression and clinical data were obtained from public databases and an independent local cohort. Kaplan-Meier and Cox regression analyses were performed to evaluate prognostic value. Differential expression analysis, pathway enrichment analysis and single-cell RNA-seq data analysis were used to explore biological functions. PPRC1 was silenced in MM cell lines using siRNA to assess its effects on cell survival and oxidative phosphorylation. RESULTS: PPRC1 was significantly upregulated in MM and was associated with advanced disease stage and poor overall survival. Multivariate Cox analysis identified PPRC1 as an independent prognostic factor. A nomogram incorporating PPRC1 and revised-ISS improved survival prediction. Functional analyses revealed that PPRC1 was positively correlated with oxidative phosphorylation and oncogenic signalling pathways. A potential connection between PPRC1 expression and immune cell infiltration was observed. PPRC1 knockdown inhibited cell proliferation, induced cell cycle arrest and apoptosis and impaired oxidative phosphorylation in MM. CONCLUSIONS: PPRC1 acts as a prognostic biomarker and metabolic regulator in MM by sustaining mitochondrial oxidative phosphorylation. These findings highlight PPRC1 as a potential therapeutic target in MM.

Humans↗

PubMed2026

Mortality of Individuals With PRNP Variants Associated With Prion Disease in the United States, 1998-2024.

BACKGROUND AND OBJECTIVES: To characterize the survival of individuals with pathogenic PRNP variants-including to estimate annual hazards, to judge the accuracy of previously reported survival data, and to evaluate the utility of public record searches in determining vital status. METHODS: In this single-center cohort study, we gathered data on individuals who received positive antemortem PRNP genetic tests at the US National Prion Disease Pathology Surveillance Center (NPDPSC), including both diagnostic tests in symptomatic individuals, and predictive tests in asymptomatic individuals. Genetic test and autopsy results were queried from the NPDPSC database, and public record searches were conducted using online tools. RESULTS: Four hundred four individuals received positive genetic test results. Of 206 cases symptomatic at the time of genetic testing, 188 are likely now deceased based on typical disease duration for their genetic variants. Combined autopsy and public record searches in combination confirmed 174 of these deaths, for an estimated 92.6% sensitivity. We evaluated the age-dependent penetrance of the reportedly highly penetrance variants D178N and E200K and the reportedly low-penetrance variant V210I. Among 99 initially asymptomatic individuals with the pathogenic E200K variant, more than 936 person-years of follow-up, 18 deaths were observed, significantly fewer than 27.4 expected according to life tables based on retrospective data. The age-dependent penetrance of E200K calculated from these longitudinal data was significantly lower than that from retrospective data, with 69% penetrance by age 80 and a median age at death of 75. For the pathogenic D178N variant, the median age at death was 57, which was numerically later, but not significantly different from, that seen in retrospective data. For V210I, just 2 deaths occurred, both after age 90, consistent with minimal penetrance. DISCUSSION: Our data support high penetrance of PRNP D178N and E200K variants and low penetrance of V210I. For E200K, the age at onset distribution appears to be shifted slightly later, and lifetime risk slightly lower, than previously reported. Autopsy data and public death records in combination were sensitive and concordant for determining long-term outcomes, but additional prospective data should be gathered to support future preventive trials.

Journal Article↗

PubMed2026

Lung function after randomization to metformin, lifestyle intervention or placebo in the Diabetes Prevention Program Outcomes Study (DPPOS).

INTRODUCTION: Metformin and physical activity have been suggested as beneficial for chronic lung disease; however, there are no prior randomized trials. METHODS: The Diabetes Prevention Program (DPP) was a 3-year trial that randomized 3234 individuals at risk for diabetes to metformin, lifestyle intervention or placebo. After the DPP, 88% of participants enrolled in the DPP Outcomes Study that offered lifestyle intervention to all and open-label continuation of metformin. Spirometry was performed at approximately 19 and 22 years post-randomization. Lung function measures were compared in an intention-to-treat (ITT) analysis by original randomization group. Models were unadjusted and adjusted for demographics, body size, smoking and sitting/standing at spirometry. Additional analyses tested prevalence of obstruction (FEV1/FVC <70%), restrictive pattern (FVC < LLN and FEV1/FVC ≥70%), preserved ratio impaired spirometry (PRISm: FEV1 <80% predicted, FEV1/FVC ≥70%) and symptoms (COPD Assessment Test [CAT] score ≥10). RESULTS: The 1888 participants with spirometry were a mean (±SD) age of 68.2 ± 9.3 years, 70% female and 6% currently smoked and 33% had previously smoked cigarettes. Mean follow-up time was 19.0 ± 0.8 years. The mean FEV1 was 2.14 ± 0.60 L, FVC 2.74 ± 0.74 L, FEV1/FVC 78.4 ± 6.4%, mean BMI was 32.4 ± 6.7 kg/m2 and 58% had diabetes. In both unadjusted and adjusted ITT analyses, randomization group was not associated with FEV1, FVC or FEV1/FVC. Likewise, rates of obstruction, restrictive pattern, PRISm or symptoms did not differ by randomization group. CONCLUSIONS: In this long-term follow-up after a randomized trial, we found no significant associations between randomization to metformin or lifestyle intervention and lung function or respiratory symptoms.

Humans↗

PubMed2026

Genome-wide identification of the HSP70 superfamily in tropical sea cucumber Stichopus monotuberculatus and their expression analysis under low-salinity stress.

Heat shock proteins (HSPs) are a group of evolutionarily conserved molecular chaperones that serve as indispensable core regulators in preserving cellular homeostasis and orchestrating organismal stress responses. The tropical sea cucumber Stichopus monotuberculatus, a high-value aquaculture species, is sensitive to fluctuations in environmental salinity-a challenge that has emerged as a critical bottleneck limiting its large-scale commercial cultivation. However, no systematic investigation has been conducted to characterize the HSP70 superfamily in S. monotuberculatus and elucidate its functional roles in salinity adaptation. In the present study, we performed a comprehensive genome-wide scan and identified 19 HSP70 superfamily genes in the S. monotuberculatus genome, with the HSP70IV subfamily showing remarkable gene expansion, containing 8 distinct copies. Phylogenetic analysis, conserved motif identification, and gene structure characterization demonstrated high evolutionary conservation within each HSP subfamily. These genes were unevenly distributed across the chromosomes of S. monotuberculatus, and prediction of cis-acting elements revealed that their upstream regulatory regions were enriched with numerous functional elements associated with stress response and immune regulation. Salinity stress experiments revealed that under severe low-salinity conditions (18‰), the expression levels of SmHSPA14L and multiple HSP70IV subfamily members were significantly elevated, while SmHYOU1D was significantly downregulated; in contrast, only subtle changes were detected in the expression of most HSP70 genes under moderate low-salinity stress (24‰). These findings strongly suggest that HSP70 genes, particularly the expanded HSP70IV subfamily, may act as key modulators in the low-salinity stress response. This work provides valuable insight into the molecular mechanisms underlying salinity adaptation in tropical sea cucumbers.

Animals↗

PubMed2026

The impact of artificial intelligence on critical thinking and clinical reasoning in health professions education: A systematic review and meta-analysis.

BACKGROUND: Critical thinking and clinical reasoning underpin healthcare professionals' ability to navigate uncertainties and deliver safe and effective care. With artificial intelligence (AI) advancement and growing adoption, AI-based educational tools are increasingly used to support these cognitive competencies' development. OBJECTIVE: To synthesize randomised and controlled clinical trials on AI-based educational tools in health professions education and examine their effects on critical thinking and clinical reasoning among health professions students. METHODS: Six electronic databases were searched from January 1, 2014 to July 28, 2025 was reviewed: PubMed, Cochrane Central Register of Controlled Trials, CINAHL, Scopus, Embase and Web of Science. Two independent reviewers performed data extraction and quality assessment using standardized JBI checklists. The GRADE approach was used to assess the certainty of evidence. Studies were pooled via random-effects meta-analyses or narrative syntheses. RESULTS: Fourteen randomised controlled trials and seven controlled clinical trials were included (n = 21). Meta-analyses revealed small to medium effect sizes for the surrogate clinical reasoning outcomes of performance-based assessment scores (SMD 0.68; 95% CI [0.38, 0.98], p-value = 0.00; I2 = 38%) and knowledge test scores (SMD 0.39; 95% CI [0.09, 0.69], p-value = 0.01; I2 = 79%). Critical thinking and clinical reasoning skills and dispositions were narratively synthesized, with majority of included studies favouring AI-based interventions but the evidence had low to very low certainty. CONCLUSION: AI-based educational interventions may improve critical thinking and clinical reasoning among health profession students, but the evidence is very uncertain. This review offers preliminary insights but does not allow identification of optimal interventions or discipline-specific recommendations due to small sample sizes and substantial intervention heterogeneity. Further research is required to draw definitive conclusions. PROTOCOL REGISTRATION: CRD42025634074.

Humans↗

PubMed2026

AI-enabled viral genomics: from virus discovery to host prediction and emerging variant forecasting.

The rapid expansion of metagenomic sequencing has generated vast repositories of viral sequence data that far outpace our capacity to interpret them using conventional approaches. Highly divergent sequences, sparse functional annotation, and taxonomically uneven sampling present fundamental challenges for reference-dependent methods, which lose sensitivity precisely for novel and understudied viruses with high public health relevance. Artificial intelligence (AI) provides a new avenue to address these challenges by enabling predictive inference from viral genomes and proteins while reducing dependence on sequence similarity. In this Review, we discuss representative advances in AI for virus discovery, taxonomic classification and functional annotation, prediction of host range and zoonotic potential, and efforts toward forecasting emerging variants. These advances are transforming viral genomics from a largely descriptive discipline into one with increasing predictive capability. We also critically assess the major challenges that constrain current approaches, including the availability of high-quality and representative datasets, rigorous model evaluation, biological interpretability and responsible governance for increasingly capable AI models.

Artificial Intelligence↗