Diverse Variables Research Articles

Visitor Number Prediction for Daegwallyeong Forest Trail Using Machine Learning

Predicting forest trail visitation is essential for sustainable management and policy development, including infrastructure planning, safety operations, and conservation. However, due to numerous informal access points and complex external influences, accurately monitoring visitor numbers remains challenging. This study applied random forest, gradient boosting, and LightGBM models with Bayesian optimization to predict daily visitor counts across six sections of the National Daegwallyeong Forest Trail, incorporating variables such as weather conditions, social media activity, COVID-19 case counts, tollgate traffic volume, and local festivals. SHAP analysis revealed that tollgate traffic volume and weekends consistently increased visitation across all sections. The impact of temperature varied by section: higher temperatures increased visitation in Kukmin Forest, whereas lower temperatures were associated with higher visitation at Seonjaryeong Peak. COVID-19 cases demonstrated negative effects across all sections. By integrating diverse variables and conducting section-level analysis, this study identified detailed visitation patterns and provided a practical basis for adaptive, section- and season-specific management strategies. These findings support flexible measures such as seasonal staffing, congestion mitigation, and real-time response systems and contribute to the advancement of data-driven regional tourism management frameworks in the context of evolving nature-based tourism demand.

Using DAP-RPA Point Cloud-Derived Metrics to Monitor Restored Tropical Forests in Brazil

Monitoring forest structure, diversity, and biomass in restoration areas is both expensive and time-consuming. Metrics derived from digital aerial photogrammetry (DAP) may offer a cost-effective and efficient alternative for monitoring forest restoration. The main objective of this study was to use metrics derived from digital aerial photogrammetry (DAP) point clouds obtained by remotely piloted aircraft (RPA) to estimate aboveground biomass (AGB), species diversity, and structural variables for monitoring restored secondary tropical forest areas. The study was conducted in three active and one passive forest restoration systems located in a secondary forest in Sergipe state, Brazil. A total of 2507 tree individuals from 36 plots (0.0625 ha each) were identified, and their total height (ht) and diameter at breast height (dbh) were measured in the field. Concomitantly with the field inventory, the plots were mapped using an RPA, and traditional height-based point cloud metrics and Fourier transform-derived metrics were extracted for each plot. Regression models were developed to calculate AGB, Shannon diversity index (H′), ht, dbh, and basal area (ba). Furthermore, multivariate statistical analyses were used to characterize AGB and H′ in the different restoration systems. All fitted models selected Fourier transform-based metrics. The AGB estimates showed satisfactory accuracy (R2 = 0.88; RMSE = 31.2%). The models for H′ and ba also performed well, with R2 values of 0.90 and 0.67 and RMSEs of 24.8% and 20.1%, respectively. Estimates of structural variables (dbh and ht) showed high accuracy, with RMSE values close to 10%. Metrics derived from the Fourier transform were essential for estimating AGB, species diversity, and forest structure. The DAP-RPA-derived metrics used in this study demonstrate potential for monitoring and characterizing AGB and species richness in restored tropical forest systems.

Source-specific dynamics of organic micropollutants in combined sewer overflows.

Combined sewer overflows (CSOs) discharge organic micropollutants (MPs) into open water bodies, posing potential environmental threats. Knowledge of the numbers, sources, and dynamics of MPs during CSOs is scarce but crucial for assessing their impact and developing mitigation strategies. To shed light on the dynamics of dissolved organic MPs in CSOs, we conducted high-temporal-resolution sampling (10 min composite samples) followed by liquid chromatography high-resolution mass spectrometry analysis, both target (60 substances) and nontarget, at two CSO sites in a small [17 hectares reduced (hared)] and a large (368 hared) catchment for over 10 events each. We observe similar patterns among indoor substances in the large catchment and among tire-associated compounds in both catchments, indicating source-specific behavior. Due to high and diverse concentration variability, no temporal correlations were found among indoor substances in the small catchment or among pesticides in either catchment. A random forest classifier was applied to assign nontarget time series to indoor and road sources in the large catchment. The results indicate that CSOs discharge several thousand substances from indoor sources, followed by a few hundred from outdoor sources with continuous leaching. These high numbers substantially surpass the scope of traditional target lists and underscore the importance of broad-spectrum screening methods when assessing MP contamination.

Agent-Based Model for Proportionality Assessment in Military Operations

The proportionality assessment is a fundamental principle and a critical consideration in military operations. It involves weighing the anticipated military advantage of a military action against the potential for collateral damage, ensuring that the harm inflicted on civilians and civilian objects is not excessive in relation to the intended military gains. This process is inherently complex, requiring decision-makers to navigate uncertain and dynamic operational environments while integrating diverse variables, such as the operational context, available intelligence, and the evolving nature of conflict. To explore and better understand this decision-making process, this research introduces a novel Agent-Based Model (ABM) designed specifically to model and simulate proportionality assessment in military operations. The model proposed captures the interactions between decision-makers, environmental variables, and operational factors, providing a dynamic platform for analysing complex proportionality scenarios. By modelling these interactions and the underlying behaviour of this assessment process, this Artificial Intelligence (AI) model enables the simulation of diverse operational contexts, offering valuable insights into the decision-making process. Through this approach, this research contributes to the ongoing development of responsible and trustworthy AI models that enhance the understanding and evaluation of proportionality in military operations, supporting the creation of more informed and ethical operational strategies.

A Machine Learning-Based Assessment of Proxies and Drivers of Harmful Algal Blooms in the Western Lake Erie Basin Using Satellite Remote Sensing

The western region of Lake Erie has been experiencing severe water-quality issues, mainly through the infestation of algal blooms, highlighting the urgent need for action. Understanding the drivers and the intricacies associated with algal bloom phenomena is important to develop effective water-quality remediation strategies. In this study, the influences of multiple bloom drivers were explored, together with Harmonized Landsat Sentinel-2 (HLS) images, using the datasets collected in Western Lake Erie from 2013 to 2022. Bloom drivers included a group of physicochemical and meteorological variables, and Chlorophyll-a (Chl-a) served as a proxy for algal blooms. Various combinations of these datasets were used as predictor variables for three machine learning models, including Support Vector Regression (SVR), Extreme Gradient Boosting (XGB), and Random Forest (RF). Each model is complemented with the SHapley Additive exPlanations (SHAP) model to understand the role of predictor variables in Chl-a estimation. A combination of physicochemical variables and optical spectral bands yielded the highest model performance (R2 up to 0.76, RMSE as low as 8.04 µg/L). The models using only meteorological data and spectral bands performed poorly (R2 < 0.40), indicating the limited standalone predictive power of meteorological variables. While satellite-only models achieved moderate performance (R2 up to 0.48), they could still be useful for preliminary monitoring where field data are unavailable. Furthermore, all 20 variables did not substantially improve model performance over models with only spectral and physicochemical inputs. While SVR achieved the highest R2 in individual runs, XGB provided the most stable and consistently strong performance across input configurations, which could be an important consideration for operational use. These findings are highly relevant for harmful algal bloom (HAB) monitoring, where Chl-a serves as a critical proxy. By clarifying the contribution of diverse variables to Chl-a prediction and identifying robust modeling approaches, this study provides actionable insights to support data-driven management decisions aimed at mitigating HAB impacts in freshwater systems.

Exploring organizational characteristics connected with ethics institutionalization

Purpose This study aims to identify organizational characteristics associated with different levels of ethics institutionalization and to understand how they contribute to the development of robust ethical frameworks. By examining these characteristics in an integrated manner, the research offers deeper insights into the organizational factors that support the establishment of strong ethical practices. Design/methodology/approach The study identifies key attributes connected with ethics institutionalization using a complex of diverse background variables such as organizational size, regional location, industry, ownership type, legal form, organizational age, profitability and professional association membership. A two-step cluster analysis and Chi-square tests were conducted on a representative sample of 1,295 Slovak organizations, stratified by company size and region. Findings In the subset of micro/small organizations, high ethics institutionalization is linked with private foreign ownership, the construction industry and company location in Western Slovakia. For medium/large companies, ethics institutionalization is connected with foreign ownership, presence in the capital city region, joint stock structure, association membership and profitability. Research limitations/implications This study was limited to organizations operating within Slovakia, which may impact the generalizability of findings to other regions. In addition, the study focused on a defined set of organizational characteristics, potentially overlooking other factors that could influence ethics institutionalization. Practical implications This research highlights which segments of businesses may benefit most from targeted ethical policies, offering managers a foundation for more effective decision-making in organizational ethics. The study’s findings can help policymakers identify specific business segments or company types needing additional regulatory guidance or support. Social implications As businesses implement more robust ethical frameworks, they contribute to raising standards across the market, promoting fair practices and reducing the occurrence of unethical behavior in business, which can have broad societal benefits. Originality/value This research advances theoretical and practical understanding of organizational ethics by examining a range of organizational characteristics, rarely explored in conjunction, with respect to ethics institutionalization. The uniqueness of this study also lies in its focus on one of the Central and Eastern European countries, a context that remains understudied in business ethics, providing valuable perspectives on a posttransformation business environment.

Association of chronotype, breakfast habits, and sleep quality with BMI-for-Age in adolescents

Adolescence is prone to nutritional imbalances. Research in SMPN 18 Surakarta showed that 50% of adolescents were thinned, 10% were overweight, and 3,3% were obese. This study aimed to analyze whether chronotype, breakfast habits, and sleep quality contribute to malnutrition among adolescents. This research was conducted in Surakarta City between November 2024-January 2025 and used a cross-sectional as the design with multistage sampling. The Lemeshow Formula showed that the minimum sample size was 106. The instruments used were MEQ, Breakfast Habits, and PSQI questionnaires as well as body weight and height measurements. Kolmogorov-Smirnov test showed that the data is normally distributed. Pearson used as the bivariate test and multiple linear regression as the multivariate test. There was no correlation between chronotype (p = 0,900; r = 0,011), breakfast habits (p = 0,298; r = 0,087), and sleep quality (p = 0,420; r = 0,067) and BMI-for-age. There was no correlation between all independent variables and the dependent variable (F = 0,585). This research concluded that there is no correlation between chronotype, breakfast habits, and sleep quality with BMI-for-age in adolescents. However, further research with more diverse subjects and lifestyle variables is required to confirm these findings.

Assessing environmental and economic dynamics in the EU agri-food sector: The impact of imports through a BVAR analysis

The study investigates the environmental impact of the EU agri-food sector, focusing on emissions per capita and their relationship with economic growth through a Bayesian Vector Autoregression (BVAR) framework. It reveals that greening efforts in the sector have not been matched by sufficient economic growth, challenging the Environmental Kuznets Curve (EKC) hypothesis. Despite progress in sustainable practices, economic expansion has fallen short of offsetting environmental costs, with imports playing a critical role. The Carbon Border Adjustment Mechanism (CBAM) under the EU Green Deal highlights the need to address trade-related emissions. The study calls for future research to develop a comprehensive index incorporating diverse variables to better assess sustainability efforts in the agri-food sector.

A CASE REPORT OF POLAND SYNDROME TREATED FOR SYNDACTYLY RELEASE AT KHYBER TEACHING HOSPITAL PESHAWAR

Background: Poland syndrome (PS) is a rare congenital condition characterized by unilateral absence or underdevelopment of the pectoralis major muscle, often accompanied by ipsilateral hand anomalies such as syndactyly. The syndrome is more common in males, predominantly affecting the right side. Severity and associated anomalies vary widely, making early diagnosis and appropriate intervention essential. Case Presentation: We report a case of a 16-year-old right-hand dominant male from a rural area of Peshawar, presenting with congenital webbing (syndactyly) of the index, middle, and ring fingers of the right hand. Clinical examination revealed an absent right pectoralis major muscle. Due to financial constraints, confirmatory imaging (CT scan) was not performed. The patient underwent successful surgical release of syndactyly between the middle and ring fingers. A second-stage release of the index and middle fingers was planned for a later date. There was no family history of PS, and no cardiac or other systemic anomalies were detected. Discussion: Poland syndrome exhibits diverse phenotypic variability. This case aligns with the classical presentation in males, involving right-sided chest wall and upper limb anomalies. The absence of systemic involvement and mild severity allowed functional and aesthetic improvement via surgical intervention. Literature suggests vascular disruption during embryogenesis as a likely pathophysiological mechanism. Timely surgical correction in such patients enhances hand function and cosmetic outcomes, especially in cases with syndactyly. Conclusion: This case highlights a mild variant of Poland syndrome involving functional impairment due to syndactyly, treated successfully at a tertiary care hospital in Pakistan. It emphasizes the importance of clinical evaluation for timely diagnosis and intervention, especially in resource-limited settings where imaging may not be feasible. To our knowledge, this is the first reported case of PS managed surgically in our institution.

Ethnic Diversity and Monitoring Effectiveness of the Board: Evidence from Banks

Synopsis The research problem We investigate the effect of ethnic diversity on the reporting quality of U.S. banks. Motivation Although the representation of ethnic minorities in the U.S. boards has increased recently, only a few studies investigated its effect on the board’s monitoring effectiveness. The test hypotheses An ethnically diverse board has a higher monitoring performance in the form of timelier loan loss provision (LLP) recognition. Target population The U.S. commercial banking sector covers the period 1996–2017. Adopted methodology Our main analysis used a fixed effect estimator. We address endogeneity concerns by using bank-fixed effects, CEO-fixed effects, and employing propensity-score-matched and entropy-balanced samples in additional tests. We use LLP, the main accrual in banks, as our measure for financial reporting quality. Our main independent variables are the ethnic diversity of the board and the ethnic diversity of the audit committee. Our ethnic diversity of the board variable is the percentage of independent non-Caucasian directors on the board. Analyses First, we regress LLP on our ethnic diversity variable, controlling for various board characteristics, CEO attributes, and the quality of banks’ information environment. We also extend our analyses to examine the effect of ethnic diversity of the audit committee on LLP timeliness. Finally, using accounting- and market-based measures of risk, we investigate whether bank risk moderates the association between ethnic diversity and LLP timeliness. Findings Our findings indicate that ethnically diverse boards provide more effective monitoring, reflected by higher earnings quality in the form of timelier LLP reporting. We also find that diverse boards are only associated with timelier LLP reporting in high-risk banks, indicating that ethnically diverse boards become more risk averse during periods of financial distress. In light of the recent increased levels of ethnic diversity in U.S. banks and the opaque financial reporting environment, our study provides evidence that ethnically diverse boards are better monitors than more homogenous ones.

Temporal and Spatial Changes in Water Quality and Phytoplankton Populations in the Lower St. Johns River, Florida.

The St. Johns River (SJR) is an ecologically and economically important estuarine river system undergoing extensive anthropogenic change. In this study, water quality parameters (dissolved oxygen, temperature, salinity, pH, hardness, alkalinity, ammonia-N, nitrate-N, and nitrite-N) and a suite of metals (cadmium, copper, lead, nickel, silver, and zinc) were measured in water samples collected from eight sites in the lower SJR from 2019 to 2022. This project was continued from previous work that documented these parameters in the river from 2017 to 2019. Aquatic communities such as phytoplankton can be indicative of river health; therefore, phytoplankton were also collected from each site, and the diatom component was identified. The total number of taxa in each sample ranged from 60 to 190, with 25 taxa accounting for the majority (64%). Similar to water quality, seasonal fluctuations in phytoplankton abundance and diversity were observed, with an increased relative abundance of Skeletonema costatum and Skeletonema subsalsum in times of lowered diversity. Furthermore, decreased phytoplankton diversity correlated with increased metal concentrations in the lower SJR. Multivariate analyses highlighted significant interactions among phytoplankton diversity and water quality variables. Significant parameters affecting phytoplankton biodiversity included salinity, pH, temperature, copper hazard quotient, and the nickel hazard quotient. This study provides new information about the impact of human disturbance on biotic communities and the complexity in predicting population changes.

Pengaruh Gender Diversity, Kompensasi Bonus, dan Komisaris Independen terhadap Manajemen Laba

The purpose of this study is to determine the effect of gender diversity, stock bonus compensation, and financial expertise of independent commissioners on earnings management in manufacturing companies in Indonesia for the period 2018-2022. This study uses secondary data obtained through the annual financial reports of manufacturing companies listed on the IDX for the period 2018-2022. This study is a quantitative study with sampling using the purposive sampling method and obtaining 135 research sample data. The analysis method used is multiple linear regression analysis using SPSS software version 27. The results of this study indicate that the Board of Directors Gender Diversity variable has a significant positive effect on earnings management. The Stock Bonus Compensation variable has a significant negative effect on earnings management. Meanwhile, the Independent Commissioner Financial Expertise variable has a positive but insignificant effect on earnings management.

Predictive value of machine learning for in-hospital mortality risk in acute myocardial infarction: A systematic review and meta-analysis.

Predictive value of machine learning for in-hospital mortality risk in acute myocardial infarction: A systematic review and meta-analysis.

Feasibility and utility of ecological momentary assessment to measure mental health issues in perinatal women: Scoping review.

Feasibility and utility of ecological momentary assessment to measure mental health issues in perinatal women: Scoping review.

Predicting thyroid cancer recurrence using supervised CatBoost: A SHAP-based explainable AI approach

Recurrence prediction in well-differentiated thyroid cancer remains a clinical challenge, necessitating more accurate and interpretable predictive models. This study investigates the use of a supervised CatBoost classifier to predict recurrence in well-differentiated thyroid cancer patients, comparing its performance against other ensemble models and employing Shapley Additive Explanations (SHAP) to enhance interpretability. A dataset comprising 383 patients with diverse demographic, clinical, and pathological variables was utilized. Data preprocessing steps included handling values and encoding categorical features. The dataset was split into training and testing sets using a 70:30 ratio. Model performance was evaluated using accuracy and area under the receiver operating characteristic curve. A comparative analysis was conducted with other ensemble methods, such as Extra Trees, LightGBM, and XGBoost. SHAP analysis was employed to determine feature importance and assess model interpretability at both the global and local levels. The supervised CatBoost classifier demonstrated superior performance, achieving an accuracy of 97% and an area under the receiver operating characteristic curve of 0.99, outperforming competing models. SHAP analysis revealed that treatment response (SHAP value: 2.077), risk stratification (SHAP value: 0.859), and lymph node involvement (N) (SHAP value: 0.596) were the most influential predictors of recurrence. Local SHAP analyses provided insight into individual predictions, highlighting that misclassification often resulted from overemphasizing a single factor while overlooking other clinically relevant indicators. The supervised CatBoost classifier demonstrated high predictive performance and enhanced interpretability through SHAP analysis. These findings underscore the importance of incorporating multiple predictive factors to improve recurrence risk assessment. While the model shows promise in personalizing thyroid cancer management, further validation on larger, more diverse datasets is warranted to ensure robustness.

Harnessing Geospatial Artificial Intelligence and Deep Learning for Landslide Inventory Mapping: Advances, Challenges, and Emerging Directions

Recent advancements in artificial intelligence (AI) and deep learning enable more accurate, scalable, and automated mapping. This paper provides a comprehensive review of the applications of AI, particularly deep learning, in landslide inventory mapping. In addition to examining commonly used data sources and model architectures, we explore innovative strategies such as feature enhancement and fusion, attention-boosted techniques, and advanced learning approaches, including active learning and transfer learning, to enhance model adaptability and predictability. We also highlight the remaining challenges and potential research directions, including the estimation of more diverse variables in landslide mapping, multimodal data alignment, modeling regional variability and replicability, as well as issues related to data misinterpretation and model explainability. This review aims to serve as a useful resource for researchers and practitioners, promoting the integration of deep learning into landslide research and disaster management.

A Multi-Feature Stock Index Forecasting Approach Based on LASSO Feature Selection and Non-Stationary Autoformer

The Chinese stock market, one of the largest and most dynamic emerging markets, is characterized by individual investor dominance and strong policy influence, resulting in high volatility and complex dynamics. These distinctive features pose substantial challenges for accurate forecasting. Existing models like RNNs, LSTMs, and Transformers often struggle with non-stationary data and long-term dependencies, limiting their forecasting effectiveness. This study proposes a hybrid forecasting framework integrating the Non-stationary Autoformer (NSAutoformer), LASSO feature selection, and financial sentiment analysis. LASSO selects key features from diverse structured variables, mitigating multicollinearity and enhancing interpretability. Sentiment indices are extracted from investor comments and news articles using an expanded Chinese financial sentiment dictionary, capturing psychological drivers of market behavior. Experimental evaluations on the Shanghai Stock Exchange Composite Index show that LASSO-NSAutoformer outperforms the NSAutoformer, reducing MAE by 8.75%. Additional multi-step forecasting and time-window analyses confirm the method’s effectiveness and stability. By integrating multi-source data, feature selection, and sentiment analysis, this framework offers a reliable forecasting approach for investors and researchers in complex financial environments.

AN INTELLIGENT FRAMEWORK FOR HYBRID NANOFLUID FLOW BETWEEN TWO CONCENTRIC CYLINDERS BASED ON DEEP NEURAL NETWORKS

This study investigates the intricate dynamics of hybrid nanofluid (HNF) flow within the annular space between two concentric cylinders. The hybrid nanofluid consists of a precise blend of ethylene glycol (80%) and water (20%) combined with copper oxide (CuO) and silver (Ag) nanoparticles. This research aims to delineate the intricate flow, temperature, and pressure profiles of the hybrid nanofluid influenced by the rotation of the outer cylinder, with the inner cylinder remaining stationary, mitigating joule heating and fluid velocity through the application of a radial magnetic field. A unique deep neural network (DNN) technique enhances model prediction accuracy by combining diverse influencing variables. Numerical simulations meticulously examine the distribution of fluid velocity, temperature, and pressure, corroborated by measurements. As the strength of the magnetic field increases, fluid velocity decreases, whereas the Brinkman number and magnetic field intensity exhibit a positive correlation with fluid temperature. Intense magnetic fields diminish fluid velocity and elevate pressure. Magnetic properties and Brinkman numbers substantially influence temperature gradients within the system. The model effectively replicates fluid dynamics conditions owing to the strong connection between projected and actual values. This work investigates the flow characteristics of hybrid nanofluids within concentric cylinders, focusing on nanoparticle concentration, magnetic field parameters, and Brinkman numbers. The results elucidate fluid dynamics in intricate systems, establishing a foundation for subsequent research and technological applications.

A 14-year prospective cohort study of type 2 diabetes development in Dutch healthy adults of South Asian origin: risk factors and the association with metabolic syndrome and HOMA-IR.

Type 2 Diabetes (T2D) imposes a disproportionate burden on the South Asian population. Their phenotype is characterized by heightened insulin resistance, even in individuals without overt T2D. Commonly used screening tools underestimate the T2D incidence in this population. The Metabolic syndrome (MetS) and the Homeostatic Model Assessment for Insulin Resistance (HOMA-IR) are indicators of insulin resistance; however, their predictive value for the development of T2D remains unexplored. Among 698 initially enrolled healthy South Asian adults aged 30 to 65 in a Rotterdam-based cardiovascular disease prevention study, 270 participants were included after a 14-year follow-up. At baseline, an extensive history, physical examination, and metabolic screening were taken. A follow-up assessment of incident T2D was conducted. Multivariable logistic regression models calculated odds ratios (ORs) for MetS, its components, and HOMA-IR and adjusted for confounders. 33 (12.2%) of participants developed T2D. The presence of MetS at baseline showed an adjusted OR of 2.6, (95% confidence interval (CI) 1.2-5.7, p = 0.02) for incident T2D. Fasting plasma glucose was the most strongly associated component of MetS (OR 3.0, CI 1.1-8.6, p = 0.04) HOMA-IR was also associated and showed an OR of 1.2 per point increase (CI 1.0-1.4, p=0.05). MetS and FPG were the most important predictors of T2D development in this South Asian cohort. These results underscore the value of diverse variables in T2D detection and give insight into which screening tools for T2D prediction should be used in this high-risk population.

Genome-wide expression in human whole blood for diagnosis of latent tuberculosis infection: a multicohort research.

Tuberculosis (TB) remains a significant global health challenge, necessitating reliable biomarkers for differentiation between latent tuberculosis infection (LTBI) and active tuberculosis (ATB). This study aimed to identify blood-based biomarkers differentiating LTBI from ATB through multicohort analysis of public datasets. We systematically screened 18 datasets from the NIH Gene Expression Omnibus (GEO), ultimately including 11 cohorts comprising 2,758 patients across 8 countries/regions and 13 ethnicities. Cohorts were stratified into training (8 cohorts, n = 1,933) and validation sets (3 cohorts, n = 825) based on functional assignment. Through Upset analysis, LASSO (Least Absolute Shrinkage and Selection Operator), SVM-RFE (Support Vector Machine Recursive Feature Elimination), and MCL (Markov Cluster Algorithm) clustering of protein-protein interaction networks, we identified S100A12 and S100A8 as optimal biomarkers. A Naive Bayes (NB) model incorporating these two markers demonstrated robust diagnostic performance: training set AUC: median = 0.8572 (inter-quartile range 0.8002, 0.8708), validation AUC = 0.5719 (0.51645, 0.7078), and subgroup AUC = 0.8635 (0.8212, 0.8946). Our multicohort analysis established an NB-based diagnostic model utilizing S100A12/S100A8, which maintains diagnostic accuracy across diverse geographic, ethnic, and clinical variables (including HIV co-infection), highlighting its potential for clinical translation in LTBI/ATB differentiation.