Topological features of lithium disilicate glass-ceramics uncovered through materials informatics.

Transit times link pollution sources to drinking water quality in a "One Water" system.

Background: Diabetes Mellitus (DM) is a primary non‑communicable disease with a rising global burden. In Indonesia, inpatient admissions for diabetic complications are increasing, placing pressure on hospital resources. Reliable forecasting of patient visits is crucial for planning manpower, bed capacity, and supply chains. Objective: To model and forecast monthly inpatient visits for diabetes mellitus at RSUD Banjar for 2025–2027 using Autoregressive Integrated Moving‑Average (ARIMA) time‑series models. Methods: A retrospective time‑series study analyzed monthly inpatient visits for diabetes mellitus, both without and with complications, recorded from January 2020 to December 2024 (60 observations per series). Time‑series plots were inspected to identify trends or seasonality. The augmented Dickey–Fuller test was used to assess stationarity. The Auto Correlation Function (ACF) and Partial Autocorrelation Function (PACF) were reviewed to specify candidate ARIMA models. Missing observations were not present; potential outliers were screened using boxplots and retained because they represented actual clinical events. Candidate models were compared using the Akaike Information Criterion (AIC), the Bayesian Information Criterion (BIC), and the Root Mean Squared Error (RMSE). Residuals were examined using the Ljung–Box Q test to verify independence. Forecasts with 95% confidence intervals were generated for 2025–2027. Results: The uncomplicated DM series comprised ≈ 240 inpatient visits, whereas the complicated series comprised ≈ 320 visits. An ARIMA (1,0,1) model was selected for uncomplicated DM, and an ARIMA (1,1,0) model for complicated DM, based on AIC/BIC and residual diagnostics. Both models produced white‑noise residuals (Ljung–Box p>0.05). Forecasts suggested gradual increases for both series; uncomplicated visits were expected to rise from about 85 in 2025 to 95 in 2027, and complicated visits from 105 to 115. The upward trends were small and not statistically significant. Forecast confidence intervals indicated a margin of error of approximately ± 10% of the predicted values. Conclusion: ARIMA models provided reasonable short‑term forecasts of inpatient visits for diabetes at RSUD Banjar. The predicted increases underscore the need for proactive planning of staffing, bed capacity and procurement. Future work should explore ARIMAX or hybrid ARIMA–machine‑learning models to improve predictive accuracy.

ABSTRACT This paper develops a multivariate ‐mixture model‐based semi‐supervised clustering methodology for datasets with incomplete records. Specifically, we consider the case where not all features are always observed, as well as the case where label information for some of the records is available, where the interest is in grouping all of them. Our modeling allows for constraints on the shape, size, and orientation of the scale matrices in the component mixtures, and develops a fast alternating expected conditional maximization algorithm for parameter estimation in the semi‐supervised framework that crucially includes the setup where not all classes in the dataset necessarily have representation in the labels. The total number of groups is assessed using Bayesian information criterion. Our approach is evaluated on simulated datasets of different clustering complexity as well as amounts and structure in the unobserved parts of the records or labels. The methodology is applied to further characterize fraudulent and legitimate credit card transactions, and also to categorize incidence and severity in heart disease patients. The publicly available R package MixtClust implements our methods.

Abstract JWST observations have revealed an overabundance of bright galaxies at z ≥ 9, creating apparent tensions with theoretical predictions within standard ΛCDM cosmology. We address this challenge using a semiempirical approach that connects dark matter halos to the observed UV luminosity through a physically motivated double power-law star formation efficiency (SFE) model as a function of the halo mass and redshift, and perform a joint Bayesian analysis of luminosity functions spanning z = 4–16 using combined Hubble Space Telescope and JWST data. Through systematic model comparison using information criteria (Akaike, Bayesian, and deviance information criteria), we identify the optimal framework requiring redshift evolution only in the low-mass slope parameter α ( z ) while maintaining other SFE parameters constant. Our best-fitting model achieves excellent agreement with observations using modest, constant UV scatter σ UV = 0.32 dex–significantly lower than the ≳1.3 dex values suggested by previous studies for z > 13. This reduced scatter requirement is compensated by strongly evolving star formation efficiency, with α increasing toward higher redshifts, indicating enhanced star formation in low-mass halos during the cosmic dawn. The model also successfully reproduces other important observational diagnostics, such as effective galaxy bias and cosmic star formation density, consistently across the full redshift range. Furthermore, model predictions are consistent up to a redshift of z ∼ 20. Our results demonstrate that JWST’s early galaxy observations can be reconciled with standard cosmology through the interplay of modest stochasticity and evolving star formation physics, without invoking extreme burstiness or exotic mechanisms.

HIV-1 plasma viral load decays in a biphasic manner during antiretroviral therapy (ART). It was hypothesized that this is due to infection of different cell types, namely CD4+ T cells and macrophages. We studied this possibility directly by modeling the decay of HIV-1 in humanized mice. We utilized previously published data from humanized T-cell only mice (TOM) and myeloid-only mice (MOM) infected with HIV-1 and treated with a potent ART regimen. Viral load decay dynamics were modeled using either a single or a biexponential decay fitted using nonlinear mixed effects techniques. Fits were compared using the corrected Bayesian information criterion (BICc). In TOM, the biphasic model was significantly better than a single-phase decay model (ΔBICc ≈ 16) despite additional parameters. In MOM, the biphasic decay was statistically better, but there was substantial uncertainty because the virus goes below detection very fast. The first-phase half-life was consistent between groups (1.2 days in MOM and 1.3 days in TOM) and similar to the half-life estimated in human infection. The second-phase decay in these mice was minimal likely due to low initial viral loads. Additional analyses with mice containing both CD4+ T cells and macrophages or X4-tropic virus-infected MOM mice confirmed the biphasic pattern, demonstrating the robustness of this result. The biphasic decline in HIV-1 occurs, even with only CD4+ T cells, refuting the hypothesis that distinct cell populations (CD4+ T cells and macrophages) drive each decay phase. These findings support an alternative model in which the observed dynamics arise from intrinsic properties of the viral infection lifecycle rather than from cellular compartmentalization.IMPORTANCEIt is well known that when antiretroviral therapy is started in people infected with HIV, the decay of virus in the periphery is biphasic early on (followed by other slower phases). One possibility for this pattern of decay is infection of two different types of cells (suggested previously to be CD4+ T cells and macrophages), with different turnovers giving rise to the biphasic decline. We addressed this issue directly in a humanized mouse model of HIV, taking advantage of mice reconstituted with just T cells and treated with antiretroviral drugs. We found that the observed decay is biphasic, which eliminates the hypothesis that the biphasic decline is due to the co-existence of the two types of cells. It is possible that integration dynamics, as we previously proposed, are responsible for the observed biphasic decline.

This research aims to model the changes over time regarding the data on the number of breeds (thousands piece) of native sheep breeds (Karaman, İvesi, Dağlıç, Kıvırcık, Sakız, Karakaya, Ramlıç, Pırlak, Pırıt, Polatlı, Karya, Asaf, Malya, Bafra, Menemen, Tahirova, Koçeri, Zom, Karakaş, Eşme, Norduz, Gökçeada, Acıpayam, Hemşin, Herik, Tuj, Karagül) published by Turkish Statistical Institute (TUIK) between 2000-2023. In the study, these data were estimated using 12 different models (Richard, Gompertz, Gamma, Orskov, McNally, Brody, Weibull, Quadratic, Wood, Wilmink, Cubic Piecewise, Cubic). The accuracy of the models was evaluated using statistical measures such as mean square error, adjusted coefficient of determination, accuracy and bias factors, Akaike information criterion, adjusted Akaike information criterion and Bayesian information criterion. The results showed that the Wilmink model provided the most appropriate prediction model by obtaining the lowest values. These findings emphasize the importance of correct model selection and it is recommended that these models be examined in more detail with longer-term data in future studies

Increasing consumption of plant-based alternatives is promoted to reduce the environmental impact of food systems, yet adoption remains limited. The aim of this study was to identify distinct consumer segments and examine differences in their perceptions, consumption habits, and trial intentions concerning plant-based dairy alternatives (PBDAs). Conceptually, it advances PBDAs segmentation by jointly incorporating pro-dairy justifications, avoidance of animal-origin considerations, and self-reported PBDAs familiarity, capturing psychological defence mechanisms alongside knowledge-related influences on adoption. Data were collected in a nationwide cross-sectional CAWI survey of 1220 Polish adults responsible for household food purchasing, stratified and quota-matched by gender, age, region, and settlement size. Factor analysis of the segmenting variables was conducted using principal component analysis with varimax rotation, followed by two-step cluster analysis. Alternative cluster solutions were compared using the Bayesian Information Criterion based on the log-likelihood (BIC-LL). The selected five-cluster solution showed acceptable to good clustering quality, as indicated by silhouette-based measures of cohesion and separation. Given the cross-sectional CAWI design and reliance on self-reported measures, the findings do not allow causal inference and should be interpreted as context-specific to the Polish, dairy-centric food culture. Cluster analysis identified five segments that differed in PBDA-related beliefs, product image evaluations, consumption patterns, and trial intentions. PBDA-oriented segments, comprising a dairy-critical segment and a dual-consumption segment, exhibited higher perceived familiarity and stronger ethical and environmental concerns and showed greater PBDA use and willingness to try new products. The dual-consumption segment reported the highest use and trial readiness. In contrast, resistant segments showed stronger dairy attachment, lower perceived familiarity, and more sceptical evaluations of PBDAs’ healthfulness, naturalness, and sensory appeal, and rarely consumed plant-based alternatives. The findings highlight substantial heterogeneity in how Polish dairy consumers perceive PBDAs, emphasising the importance of segment-specific approaches for communication and product development. Tailored strategies can help address the diverse motivations and barriers of consumers, supporting a dietary shift toward more plant-based options.

Abstract We develop a three-dimensional (3-D) magnetotelluric (MT) inversion method based on Akaike’s Bayesian Information Criterion (ABIC), which statistically determines the optimal regularization parameter that balances the prior constraints and data fitting, and therefore controls the smoothness of the final model by maximizing marginal likelihood in Bayesian statistics. To mitigate the high computational cost of calculating the ABIC indicator, we introduce a low-rank transformation, similar to that used in model-to-data space inversions, resulting in a data-space variant of ABIC. This adaptation significantly reduces computational complexity, facilitating the practical application of this statistical approach in 3-D MT inversion. Our discussion on the methodology and its application begins with a theoretical analysis of the proposed inversion method, including the underlying assumptions, the mathematical foundations of this statistical framework, a detailed derivation of the data-space ABIC formulation, and the procedure for obtaining the solution that maximizes the marginal likelihood. We then assess the method's performance using synthetic data, evaluate its stability under different inversion configurations, and benchmark it against other inversion strategies, including those based on the L-curve criterion, the cooling approach, and the original ABIC in model space. We demonstrate that the proposed inversion based on the data-space ABIC formulation yields stable performance across various configurations and offers improved objectivity compared to L-curve and cooling-based inversions. It also gains a substantial computational advantage over the original model-space ABIC formulation, achieving over a 70-fold speedup in synthetic tests while preserving statistical rigor. This facilitates a smooth transition of ABIC-based MT inversion from 2-D to 3-D and also opens the door to its broader application in other large-scale inverse problems. Finally, we briefly consider a field data application of the proposed method to the Yellowstone–Snake River Plain region using EarthScope USArray MT data to showcase its practical applicability.

BackgroundCognitive dispersion is the variability in an individual's performance across multiple neuropsychological measures. One issue that has precluded cognitive dispersion from being more widely used in clinical practice is the uncertainty over which dispersion index should be preferred, the intraindividual standard deviation (ISD) or coefficient of variation (CoV).MethodA total of 200 participants were included in this study (age 69.9 ± 6.7 years, education 13.2 ± 2.6 years, female 65%), consisting of 100 cognitively unimpaired (CU) and 100 amnestic mild cognitive impairment (MCI) participants who were age‐, sex‐, and education‐matched. For each participant, raw scores of 21 measures from 16 neuropsychological tests were converted into T‐scores based on the distribution of scores of the cohort. We calculated the (1) overall test battery mean (OTBM), the average of the T‐scores of all neuropsychological measures; (2) ISD, the SD of all T‐scores, with higher values reflecting higher cognitive dispersion; and (3) CoV, the ISD divided by OTBM, which is considered ISD adjusted for mean performance. Logistic and linear regression analyses were applied to examine the association between dispersion and CU/MCI classification and the bilateral entorhinal cortical thickness, respectively.ResultUsing Akaike information criterion (AIC) and Bayesian information criterion (BIC) as the regression model selection criteria, we found that ISD in context of OTBM, i.e., including the interaction between OTBM and ISD as an independent variable, was more effective in predicting CU/MCI classification and entorhinal cortical thickness than CoV alone (Tables 1 and 2). Specifically, at low OTBM, participants were more likely to have greater entorhinal cortical thickness if their IIV is low, but not when their IIV is high (Figure 1). Multicollinearity was not serious in our regression models (all variance inflation factors for the effects of age, OTBM, and ISD/CoV were less than 2).ConclusionTaken together, these results imply that ISD and CoV exhibit distinct relationships with outcomes in our study and should not be used interchangeably. Importantly, parsing out ISD and OTBM can give further insight into how these two constructs interact with one another, an information that is potentially lost when only CoV is used.

Context . Transit-timing variations (TTVs) are a powerful tool for detecting unseen companions in systems with known transiting exoplanets and for characterising their masses and orbital properties. Large-scale and homogeneous TTV analyses represent a valuable method to complement the demographics of planetary systems and understand the role of dynamical interactions. Aims . We present the results of a systematic TTV analysis of 423 systems covering ∼16000 transits, each with a single transiting planet first discovered by the NASA TESS mission and then confirmed or validated by follow-up studies. The primary aim of this survey is to identify the most promising candidates for dynamically active systems that warrant further investigation. Methods . Our analysis was performed in a two-stage pipeline. In the first stage, precise measurements of individual transit times are extracted from the TESS light curves for each system in a homogeneous way. In the second stage, we applied a two-tiered decision framework to classify candidates by analysing the resulting transit variations. Based on excess timing scatter ( χ mod 2 ) and the difference in Bayesian information criterion (ΔBIC) of periodic models over linear ones, the TTVs were classified as significant, marginal, or non-detections. Results . We find 11 systems with significant TTVs, five of which were announced in previous works, and ten more systems with marginal evidence in our sample. We present three-panel diagnostic plots for all the candidates, showing phase-folded light curves, the transit variations over time, and the same variations folded on the recovered TTV period. A comprehensive summary table detailing the fitted parameters and TTV significance for the entire survey sample is also provided. Conclusions . This survey constitutes the largest homogeneous TTV analysis of TESS systems to date. We provide the community with updated ephemerides and a catalogue of high-quality TTV candidates, enabling targeted follow-up observations and detailed dynamical modelling to uncover the nature of unseen companions and study system architectures.

In 1969, the International Monetary Fund (IMF) introduced Special Drawing Rights (SDR) as a financial instrument to supplement currency reserves of member states. SDR allow members to draw money in a currency of their choosing. 2025 IMF data valued SDR held by member states at over USD 660 billion. SDR are, thus, a critical financial indicator of significant potential impact on global economic stability. This study analysed SDR prices from November 2015 to October 2025. The Minimal Information Criterion and the Bayesian Information Criterion were applied to compare ARIMA models, and the ARIMA (1,1,0) emerged as the best fit. Model diagnostics confirmed that no validity assumptions were violated. Observed values were regressed against fitted values. The R^2 value was over 90%, indicating a very strong linear relationship, which is very plausible. The model was then used to forecast SDR prices for the months from November 2025 to April 2026. The findings revealed a slight decline in SDR prices in the forecasted period. These insights have a significant impact on IMF member states, investors and international economic policy makers. Keywords: Special drawing rights (SDR), autoregressive moving average (ARIMA), time series analysis, forecasting.

ABSTRACT Synthetic daily rainfall series are essential for agricultural planning, hydraulic infrastructure design and integrated water‐resource management, particularly in semi‐arid regions where observational records are often incomplete or short. This study evaluates two parametric stochastic models, first‐order Markov Chains coupled with gamma (GAM) and mixed exponential (ME) distributions, to generate synthetic rainfall series for three stations in the Agreste mesoregion of Pernambuco, Brazil. Transition probabilities for dry and rainy days were estimated up to third order and optimised via the Bayesian Information Criterion (BIC), which overwhelmingly selected the first‐order chain. A total of 1000 synthetic daily rainfall series were generated for each station to assess distributional fidelity and temporal behaviour. Both GAM and ME adequately reproduced observed daily and monthly precipitation statistics. GAM achieved closer agreement with the observed daily‐series mean precipitation (calculated over the full daily record, including dry days) and with the monthly totals, with deviations below 5% at all stations, while ME more accurately captured extreme rainfall events, reducing the “overdispersion” common in two‐parameter models. These findings demonstrate that combining a first‐order Markov occurrence model with ME is preferable when simulating heavy rainfall extremes, whereas GAM offers robust performance for average and total rainfall estimates. The complementary strengths of these distributions provide a flexible framework for synthetic rainfall generation in data‐scarce, semi‐arid environments, supporting improved hydrological risk assessment and resource planning under variable climatic conditions.

To address the issue of individual differences affecting the accuracy of non-invasive prenatal testing (NIPT), especially the pain point of low free DNA concentration in the fetus of pregnant women with high BMI, this study analyzed the correlation between the concentration of the Y chromosome in the fetus and the gestational age and BMI of pregnant women. First, samples with abnormal GC content (not 40% to 60%) were excluded through data preprocessing. Then, histograms and scatter plots were drawn to initially observe the distribution and correlation characteristics of the variables. It was found that the scatter plot was difficult to visually determine the correlation. Subsequently, a grouped regression strategy was adopted: BMI was grouped by equal frequency, gestational weeks were determined based on the Jenks natural breakpoint method, combined with variance goodness-of-fit (GVF), pseudo-F statistics, Bayesian information criterion (BIC), and entropy weight method - TOPSIS method to determine the optimal 15 groups. After taking the mean of each group of variables, a univariate linear regression model was constructed, and the significance was verified through the F-test. The results showed that the concentration of the Y chromosome was significantly positively correlated with the gestational weeks (r= 0.6338, P=0.0112), and the degree of correlation was moderate. It was extremely significantly negatively correlated with BMI (r= -0.8658, P=0.0054), and the degree of correlation was relatively strong. This study provides data support for optimizing the timing of NIPT detection and improving the accuracy of detection in pregnant women with high BMI.

This study investigates the combined effects of cognitive, psychosocial, and cross-subject academic indicators on students’ mathematics ‎achievement across 38 countries that participated in the PISA 2022 creative thinking assessment. Drawing on data from 144,446 students, ‎we employed hierarchical linear modeling to examine how reading and science proficiency, engagement in creative activities (in and out of ‎school), perseverance, curiosity, and socioeconomic status (ESCS) predict mathematical performance. The results show that reading and ‎science are robust predictors of mathematics scores. ESCS and perseverance also demonstrated consistent positive effects, while creativity ‎showed context-specific associations, positive in some clusters and negative in others. To identify latent cross-national typologies, we ap-‎plied both K-means clustering and Gaussian Mixture Modeling (GMM) to country-level aggregates. Model comparison using the Bayesian ‎Information Criterion (BIC) favored the GMM solution, which was subsequently used to group countries for multigroup structural equation ‎modeling (MG-SEM). Results revealed significant variations in predictor effects across clusters, highlighting heterogeneity in pathways to ‎mathematics success. This study contributes to comparative education research by integrating hierarchical regression and latent classification ‎techniques, offering implications for instructional design and international education policy aimed at promoting mathematical literacy across ‎diverse systems‎.

Clustering is a fundamental problem in many scientific applications. This paper introduces the concept of K-regression, which divides a random sample of size n into K clusters such that the observations within each cluster exhibit an identical linear pattern of dependence, and the observations in different clusters exhibit distinctive structures of linear dependence. We estimate the coefficients of the clustering regressions through minimizing the within cluster l 1 and l 2 loss functions. From the asymptotic perspective, the resulting estimates obtained with either the l 1 or the l 2 loss are strongly consistent and asymptotically normal. From the non-asymptotic perspective, we further explore the conditions under which the models are identifiable and the algorithms are convergent. Furthermore, we propose a tailored Bayesian Information Criterion (BIC) designed specifically for regression-based clustering. Through extensive simulations and an application to clinical trial subgroup analysis, we demonstrate the effectiveness of K-regression. Numerical results highlight that, in the presence of heterogeneity, l 1 K-regression outperforms alternative methods (including l 2 K-regression) in coefficient estimation, cluster number determination, and subgroup classification while maintaining computational efficiency. These advantages make l 1 K-regression particularly appealing for large-scale data analysis, especially when heterogeneous subpopulations are present.

Little is understood of lake browning (due to increased dissolved organic carbon; DOC) in large lakes such as the Laurentian Great Lakes. Lake browning can alter whole lake ecosystems, including decreasing exposure to damaging ultraviolet radiation (UV-B) which is strongly and selectively attenuated by DOC more so than photosynthetically active radiation (PAR). We compared the changes in UV-B and PAR transparency to DOC data collected during the ice-free seasons from 62 nearshore sites in four of the five Great Lakes from 2002 to 2022 using linear mixed effects regression models based on backwards selected Bayesian information criteria. Regionally, DOC significantly increased from 2002 to 2022 by 0.5% per year on average. DOC strongly and inversely explained the variability of UV-B and PAR transparencies, as did seasons and offshore influence on these habitats. We provide regional evidence of lake browning within the nearshore habitats of the Great Lakes as a strong contrast to the well-documented increased offshore water transparency associated with the spread of invasive dreissenid mussels

Esophagojejunal (EJ) leakage is a serious complication following total gastrectomy for gastric cancer. While several nutritional and treatment-related risk factors have been described, the role of Candida esophagitis (CE) in anastomotic failure has not been previously investigated. This retrospective cohort study included 268 patients with gastric adenocarcinoma who underwent total gastrectomy with EJ anastomosis. The study was conducted between March 2021 and March 2025 at a tertiary referral center. CE was diagnosed by histopathologic examination of proximal esophageal margins submitted during surgery. Univariable and multivariable logistic regression analyses were performed to determine predictors of EJ leakage. Best subsets variable selection using Akaike's Information Criterion (AIC) and the Bayesian Information Criterion (BIC) guided final model development. Among the 268 patients, 48 (17.9%) developed EJ leakage. Multivariable analysis identified CE (OR: 2.19, p = 0.043), hypoalbuminemia (< 3.5g/dL) (OR: 3.08, p = 0.007), BMI ≥ 25kg/m2 (OR: 3.68, p = 0.004), and administration of neoadjuvant therapy (OR: 2.34, p = 0.024) as independent predictors of EJ leakage. Additional analyses of diagnostic timing indicated that CE detected only on permanent histology (delayed treatment) was associated with higher odds of leak (adj OR 3.62; 95%CI: 1.42-9.23; p = 0.007), whereas CE detected on intraoperative frozen section was not. CE was associated with increased odds of EJ leakage after adjustment, but causality cannot be inferred from this retrospective study. The finding that delayed CE diagnosis was linked to higher leak risk suggests diagnostic timing may matter. Prospective validation of targeted esophageal assessment and timed antifungal strategies is warranted. Elevated BMI, hypoalbuminemia, and neoadjuvant therapy also contributed to higher odds of EJ leakage.

Even though mastery of suturing is a core technical skill in surgical education, existing tools for its assessment often lack psychometric validation or are not specifically designed for undergraduate training. The aim of this study was to develop and validate the Minho Suture Assessment Scale (Minho-SAS), a structured instrument to evaluate fundamental suturing competencies in medical students. The research question was whether the Minho-SAS demonstrates validity and reliability as a psychometric tool. The development process involved collaboration with multidisciplinary surgical teams and experienced practitioners to ensure content validity. Data from a cohort of medical students were utilized for psychometric evaluation. Dimensionality was assessed using parallel analysis, Bayesian information criterion, unidimensional congruence, item unidimensional congruence, explained common variance, item explained common variance and mean of item residual absolute loadings. Validity based on internal structure was assessed with Rasch model analysis and factor analysis from the tetrachoric correlation matrix. Reliability was assessed using Rasch model standard errors of measurement to obtain a conditional reliability curve and Cronbach's alpha and McDonald's omega internal consistency coefficients. Analyses supported a unidimensional structure for the Minho-SAS. The single-factor solution explained 39.96% of variance, and Rasch measures accounted for 29.15% (16.43% by persons, 12.72% by items). Residual correlations, factor loadings, and item fit statistics were within acceptable ranges. Reliability indices were satisfactory: Rasch reliability = 0.706; McDonald's omega = 0.889; Cronbach's alpha = 0.883. The Minho-SAS is a robust instrument specifically tailored for assessing fundamental suturing skills among medical students. Rasch model analysis yielded less favorable results than factor analysis, yet still acceptable. While demonstrating considerable potential, further exploration of Minho-SAS across diverse populations and educational settings is crucial to affirm its broader applicability and impact in medical education and clinical practice.

This study aimed to provide age-related osseous measures of the temporomandibular joints (TMJs) and mandibular ramus in asymptomatic children and adolescents and to develop percentile reference data for ramus height. A retrospective cross-sectional study was conducted on 133 asymptomatic participants (67 males, 66 females; age range 0-19 years) who underwent magnetic resonance imaging of the head, including the TMJs and posterior parts of the mandible. Mandibular ramus height, glenoid fossa depth, articular eminence inclination angle, and condylar dimensions (mediolateral and anteroposterior) were measured. A single observer assessed 266 TMJs, and a second observer repeated measurements for 80 TMJs to evaluate intra- and inter-observer reliability using the intraclass correlation coefficient (ICC) and Bland-Altman plots. Age-related growth curves were estimated using the Box-Cox Cole and Green (BCCG) distribution, modeling median, coefficient of variation, and skewness as cubic spline functions of age. The optimal spline degrees of freedom were selected using the Bayesian information criterion. Percentile curves were derived, and biannual lookup tables were provided. Ramus height, glenoid fossa depth, articular eminence angle, and condylar width increased with age, whereas condylar depth exhibited variable patterns during adolescence. Intra-observer reliability was excellent for all measurements. Ramus height and condylar width measurements demonstrated excellent inter-observer reliability, while glenoid fossa depth, articular eminence inclination angle, and condylar depth measurements showed good inter-observer reliability. This study provides age-related osseous measures of the TMJs and mandibular ramus in asymptomatic children and adolescents aged 0-19 years. The percentile curves for mandibular ramus height may aid in assessing growth in children with juvenile idiopathic arthritis for monitoring disease activity and treatment responses.