Search Results (317)

This study presents a novel flexible Rayleigh–exponential mixture detection model (REMDM) for estimating population abundance under line transect sampling. The proposed detection function combines a Rayleigh-type component with an exponential component to provide greater flexibility in modelling perpendicular distance data and capturing the complex detection patterns commonly observed in ecological surveys. The model exhibited smooth behaviour near the transect line and flexible tail decay, making it suitable for heterogeneous detection structures. Several statistical properties of the proposed REMDM were derived, including the probability density function, cumulative distribution function, moments, and hazard rate function. Parameters were estimated by using the maximum likelihood estimation method. The performance of the estimators is evaluated through extensive Monte Carlo simulation studies under various sample sizes and parameter settings. The simulation results indicate that the proposed estimators are consistent and efficient in terms of bias and mean squared error, with improved performance as the sample size increases. The applicability of the proposed model is demonstrated using a real perpendicular distance dataset and model performance is assessed using several goodness-of-fit measures, including the Akaike Information Criterion, Bayesian Information Criterion, Kolmogorov–Smirnov statistic, Anderson–Darling statistic, and Cramér–von Mises statistic. The results show that the REMDM provides a superior fit to several existing detection functions. In general, the proposed model offers a flexible and effective alternative for modelling detection probability and improving population abundance estimates in ecological distance sampling. Full article

Multi-criteria group decision making (MCGDM) under linguistic uncertainty remains a fundamental challenge in applied mathematics, where decision makers seldom assign crisp numerical evaluations and frequently exhibit heterogeneous risk attitudes shaped by behavioural factors. An integrated mathematical framework, hereafter PLR-3WBC (Probabilistic Linguistic Regret-driven Three-Way Bayesian Consensus), is developed to systematically integrate four methodological components that have each been individually validated in the MCGDM literature: representation of decision information with explicit probability mass on linguistic terms; quantification of decision-maker regret and rejoice psychology under linguistic uncertainty; classification of alternatives into three actionable decision regions rather than a single-valued ranking; and group consensus reaching with credal weight aggregation. Each component has demonstrated its effectiveness in its respective domain; the present framework capitalises on their complementary strengths by embedding them within a single pipeline equipped with formal guarantees, an integration that has not been previously reported. The framework integrates five methodological components: probabilistic linguistic term sets (PLTS) for information representation; the Bayesian best–worst method (BBWM) for credal criterion weighting; a regret–rejoice value function adapted to the linguistic domain for behavioural evaluation; three-way decision (3WD) thresholds derived from a loss-function model for actionable classification; and a distance-based consensus reaching process with feedback mechanism for group convergence. A case study on age-friendliness evaluation of twelve aging urban residential communities under an indicator system of five dimensions and eighteen criteria, with four expert decision makers, demonstrates that PLR-3WBC delivers an actionable three-way classification, recovers a transparent group consensus, and produces rankings broadly consistent with classical TOPSIS, VIKOR, PROMETHEE-II, and BWM-TOPSIS (Spearman rank correlation exceeding 0.97), thereby confirming that the integrated framework preserves the ordinal reliability of these established methods, while additionally delivering three outputs that arise from the methodological integration: an actionable three-way classification enabling discrete budget-aligned decisions, credal weight intervals quantifying the depth of expert agreement on criterion importance, and a behavioural reordering of borderline non-dominated alternatives that reflects the loss-averse psychology of the decision panel and would remain hidden under single-method deployment. Sensitivity analyses with respect to the regret aversion coefficient, the loss function parameters, and the consensus threshold confirm that the qualitative classification is stable across a wide parameter envelope, supporting the practical deployment of PLR-3WBC in age-friendly community renewal programmes. Full article

Background and Objectives: Patients undergoing incisional hernia repair constitute a clinically heterogeneous population with variable postoperative outcomes. Conventional risk models based on isolated risk factors may inadequately capture this complexity. This study aimed to identify data-driven clinical phenotypes and evaluate their association with surgical outcomes. Methods and Materials: A retrospective cohort of 1262 patients undergoing retromuscular incisional hernia repair (Rives–Stoppa technique) was analyzed. Unsupervised clinical phenotyping was performed using latent class analysis based on seven preoperative variables. Model selection was guided by Akaike information criterion (AIC), Bayesian information criterion (BIC), entropy, and clinical interpretability. Postoperative outcomes were compared across phenotypes. Results: Three distinct phenotypes were identified: metabolic (34.6%), structural (33.9%), and frailty (31.5%). The structural phenotype showed the highest complication (22.7%) and recurrence rates (8.6%), while the frailty phenotype had the lowest complication burden (14.6%). The metabolic phenotype was characterized by obesity and diabetes, consistent with increased wound-related morbidity. Cluster robustness was supported by internal validation metrics and sensitivity analyses. Conclusions: In this retrospective single-center cohort, distinct clinical phenotypes with different outcome profiles were identified among patients undergoing incisional hernia repair, supporting the concept that this population comprises clinically heterogeneous subgroups with distinct patterns of vulnerability. These findings should be considered preliminary and hypothesis-generating. Further external validation and prospective studies are required to determine the clinical utility of phenotype-based risk stratification. Full article

With the rapid advancement and cost reduction in high-throughput sequencing technologies, the accumulation of large-scale, three-dimensional gene expression data has surged. Consequently, effectively reducing the dimensionality of these complex datasets to extract critical biological information remains a significant challenge. Although various methods for identifying gene expression modules have been developed, most do not explicitly account for the multifactorial interactions among the temporal, spatial, and environmental dimensions. To address this limitation, we propose a novel three-dimensional triclustering technique based on a multivariate Gaussian mixture model (MVGMM) within a maximum likelihood framework. Specifically, our approach incorporates Legendre polynomials to model the temporal dynamics of gene expression and utilizes the Bayesian Information Criterion (BIC) to determine the optimal number of clusters. To further evaluate the model’s robustness against the high background noise typically present in empirical datasets (such as Arabidopsis thaliana), we conducted a rigorous sensitivity analysis by artificially injecting high-intensity Gaussian white noise into the simulated dataset. Despite severe noise interference, the global minimum of the BIC consistently remained at K = 6. Furthermore, the penalty term in the BIC successfully suppressed artificial cluster proliferation, preventing the model from fitting the noise as new functional modules. The MVGMM framework successfully recovered the predefined cluster structure in simulation studies and identified distinct expression modules in empirical Arabidopsis thaliana data. By jointly modeling temporal, spatial, and environmental variation, this study provides a statistical framework for exploring multidimensional gene expression patterns and may facilitate the identification of coordinated regulatory programs in complex biological systems. Full article

Background: Whether the fatal and non-fatal composition of aggregate pancreatitis burden has changed over time remains unclear. We assessed long-term changes in the fatal-to-non-fatal composition of aggregate pancreatitis burden using Global Burden of Disease (GBD) 2023 estimates. Methods: We conducted a systematic descriptive and trend analysis using publicly available estimates from the GBD 2023 Results Tool for incidence, prevalence, deaths, years lived with disability (YLDs), years of life lost (YLLs), and disability-adjusted life-years (DALYs) across 204 countries and territories from 1990 to 2023. Because GBD reports pancreatitis as an aggregate cause category, the analysis could not distinguish acute pancreatitis, recurrent acute pancreatitis, chronic pancreatitis, or acute exacerbations of chronic pancreatitis. Primary analyses used age-standardised rates per 100,000 population. Four burden–composition metrics were derived within each location–year stratum: the YLL:YLD ratio, YLD:DALY proportion, deaths-to-incidence ratio, and prevalence-to-incidence ratio. Temporal trends were modelled in R version 4.5, using segmented regression, with up to three joinpoints selected by a Bayesian information criterion. Results: Globally, all six age-standardised native GBD measures declined between 1990 and 2023. The age-standardised incidence rate decreased from 37.62 (95% UI 32.20–43.11) to 32.91 (28.84–37.17) per 100,000, prevalence from 93.78 (69.26–126.25) to 68.92 (52.53–90.32), deaths from 1.76 (1.49–2.16) to 1.40 (1.21–1.66), YLDs from 5.70 (2.75–9.45) to 4.34 (2.18–7.04), YLLs from 55.96 (46.50–69.72) to 43.60 (36.89–53.53), and DALYs from 61.66 (50.62–75.61) to 47.94 (40.57–58.16). However, the fatal-to-non-fatal composition changed little: the global YLL:YLD ratio was 9.82 in 1990 and 10.04 in 2023, while the YLD share of DALYs was 0.092 and 0.091, respectively. Joinpoint modelling showed fluctuation rather than a sustained shift toward disability-dominant burden: the global YLL:YLD ratio was stable until 1998, increased from 1998 to 2002 (annual percent change [APC] 1.38%, 95% CI 0.42 to 2.36), and then declined modestly thereafter (APC −0.13%, −0.20 to −0.06). Burden remained higher in males, whereas females had a greater non-fatal share of total burden (YLD:DALY in 2023: 0.134 vs. 0.073). All sociodemographic index strata remained mortality-dominant in both 1990 and 2023; low-SDI settings had the greatest fatal dominance (YLL:YLD 34.94 in 1990; 24.72 in 2023). Using a descriptive YLD:DALY ≥ 0.50 benchmark, 203 of 204 countries and territories remained below the disability-dominant threshold in both years, no country crossed from below to above this benchmark, and only Georgia moved from above to below the benchmark. Conclusions: Despite declines in global incidence, mortality, and DALY rates, the aggregate GBD pancreatitis burden remained overwhelmingly mortality-dominant from 1990 to 2023. Because GBD pancreatitis combines acute and chronic pancreatitis, this finding should be interpreted as describing the modelled aggregate pancreatitis cause category rather than proving subtype-specific mortality dominance. The intensity of fatal dominance varied by sex, SDI, region, age, and country, but a structural shift toward disability-dominant aggregate burden was not observed. Full article

Accurate short-term load forecasting is essential for reliable power system operation under increasingly nonlinear, volatile, and multi-scale load patterns. This study proposes a VMD–BayesXGB–BiLSTM hybrid forecasting framework that integrates time-series-cross-validation-based variational mode decomposition (VMD), Bayesian-optimized XGBoost (BayesXGB), and BiLSTM residual correction. First, abnormal values in the raw load and explanatory variables are detected using the $3\sigma$ criterion and corrected by cubic spline interpolation. Then, VMD parameters are selected only within the training sequence, and leakage-free VMD features are generated from historical input windows, avoiding the use of future information. BayesXGB is employed as the primary forecasting model to capture nonlinear relationships between historical load, VMD-derived multi-scale features, and external variables. Finally, a stacked BiLSTM module learns temporal patterns from historical BayesXGB predictions and residuals, and the predicted residual correction is added to the preliminary forecast. Experiments on an Australian electricity load dataset show that the proposed model achieves an RMSE of 122.1003, an MAE of 90.7386, a MAPE of 1.0269%, and an R² of 0.9921, outperforming all compared baseline models while maintaining sub-millisecond inference per sample. Full article

Road safety is a global priority, and driver behavioral factors are among its most critical determinants. Although the literature has advanced in characterizing driving styles using psychometric instruments such as the Multidimensional Driving Style Inventory (MDSI), an empirical gap persists in the simultaneous integration of latent behavioral profiles, driving exposure, and road safety outcomes, particularly in Latin American contexts and across generational extremes. This study examined the relationship between latent driving style profiles and road safety outcomes among young (18–25 years) and older (≥65 years) licensed drivers in Ecuador, while evaluating the moderating role of driving exposure. A structured survey based on the MDSI was administered to 833 active drivers, and data were analyzed using Latent Profile Analysis (LPA) and binary logistic regression. The six-profile solution was selected according to the Bayesian Information Criterion (BIC = 11,655.07), with acceptable classification quality (entropy = 0.860; minimum posterior probability = 0.808); for inferential parsimony, these profiles were subsequently consolidated into three analytically interpretable categories: Predominantly Careful, Predominantly Risky, and Distress-Reduction. The Predominantly Risky profile was significantly associated with higher odds of traffic accident involvement (OR = 2.76, 95% CI [1.55, 4.93]), whereas the Distress-Reduction profile showed substantially higher odds of receiving traffic infraction fines (OR = 4.74, 95% CI [1.69, 13.34]). The composite driving exposure index was a robust predictor across both models (accident model: OR = 2.82, 95% CI [1.60, 5.14]; fine model: OR = 1.87, 95% CI [1.29, 2.74]). In addition, a significant interaction was observed between the Predominantly Risky profile and driving exposure in the model predicting traffic infraction fines, suggesting that exposure amplified sanction risk within this behavioral category. Older drivers showed a substantially higher representation of the Distress-Reduction profile than young drivers. These findings underscore the utility of person-centered approaches for identifying heterogeneous driver configurations and for designing profile-differentiated road safety interventions; from a practical perspective, these results support the development of targeted road safety programs that integrate behavioral profile screening with exposure-based risk management for young and older drivers. Full article

Vapor Pressure Deficit (VPD) is a critical determinant of atmospheric evaporative demand and plant water stress in tropical agricultural systems. This study applied a Gaussian Mixture Model (GMM) and K-Means clustering to 36,528 hourly meteorological observations collected from Eastern Thailand between August 2021 and September 2025, with the objective of identifying distinct atmospheric moisture regimes relevant to precision irrigation management in durian cultivation. Two input configurations were evaluated: a multivariate feature space comprising air temperature, relative humidity, wind speed, solar radiation, and VPD; and a univariate input consisting of VPD alone. Model selection for GMM was guided by the Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC), while K-Means performance was assessed using the Elbow method, Silhouette Coefficient, Calinski–Harabasz Index, and Davies–Bouldin Index. For the multivariate input, GMM identified K = 7 as the optimal number of clusters, supported by the largest single-step reduction in both AIC and BIC at this transition point. For the univariate VPD input, K = 5 was selected as the most parsimonious and agriculturally interpretable solution. The seven clusters derived from the multivariate GMM were organized into four atmospheric moisture regimes, such as very low, moderate, high, and very high evaporative demand, capturing the full spectrum of diurnal and seasonal VPD variability characteristic of Eastern Thailand. The results demonstrate that GMM-based probabilistic clustering applied to multivariate meteorological inputs provides a more comprehensive characterization of atmospheric moisture dynamics than univariate or geometric clustering approaches, offering a practical framework for tiered irrigation scheduling and drought stress early warning systems in tropical fruit cultivation. Full article

The escalating frequency and intensity of extreme rainfall events driven by climate change threaten infrastructure resilience and societal safety, underscoring the urgent need for robust models to predict these events. Previous studies on the integration of Extreme Value Theory (EVT) and machine learning in modelling extreme rainfall events have not explored the use of a time-varying threshold. This study introduces a novel time-varying threshold Generalised Pareto (GP) regression tree for modelling extreme rainfall in Durban, South Africa. The proposed hybrid model combines EVT with covariate-driven regression tree partitioning, allowing the threshold to evolve dynamically with meteorological conditions. Using daily rainfall and meteorological covariate data from 1981 to 2025, the model was developed, pruned, and benchmarked against a static-threshold GP regression tree and a time-varying threshold Generalised Pareto Distribution (GPD). Evaluation based on the Bayesian Information Criterion (BIC) and log-likelihood demonstrated the superior performance of the proposed model in capturing covariate-driven heterogeneity and temporal variability of rainfall extremes. Four distinct climatic regimes with different tail behaviours and return levels were identified. This study provides the first meteorological application of a time-varying threshold GP regression tree and offers practical insights into flood risk assessment and climate resilience planning in the city of Durban. Full article

Background: Breastfeeding benefits mothers and infants, and the promotion of breastfeeding is important. Feeding strategies include exclusive breastfeeding, feeding with both human milk and formula, and exclusive formula feeding. Objectives: This study was conducted to clarify the actual situation by assuming that mixed feeding comprises several groups with different characteristics. At the same time, the study also aimed to clarify the factors associated with breastfeeding. Methods: Single-term infants without underlying disease born at Tokyo Metropolitan Hospital between 2019 and 2024 participated in this study. The distribution of formula intake among infants receiving both human milk and formula was analyzed using a Gaussian mixture model, and the optimal number of distribution components was calculated using the Bayesian information criterion. Using linear regression analysis, factors associated with formula intake were identified. Results: A total of 2628 participants (exclusive breastfeeding, 842 (32.0%); mixed feeding with human milk and formula, 1496 (56.9%); and exclusive formula feeding, 290 (11.0%)) were included in the study. Linear regression analysis showed that the factors associated with amount of formula intake were late preterm birth (coefficient 39.7, p < 0.01), maternal age (reference under 30 y, age ≥ 30 y and <35 y coefficient 6.3, p = 0.66, age ≥ 35 y and <40 y coefficient 45.5, p < 0.01, age ≥ 40 y coefficient 106.9, p < 0.01) and delivery mode (cesarean section, coefficient 53.6, p < 0.01). Conclusions: Feeding strategies involving both human and formula milk are not homogeneous, and interventions should be developed based on these differences. Moreover, several factors were found to be associated with breastfeeding, which may help promote breastfeeding. Full article

Environmental condition assessment is essential for the design of floating wind turbines, particularly when determining design sea states that balance safety and economy. The environmental contour method, typically constructed through the Inverse First Order Reliability Method combined with parametric joint distributions, is widely adopted for this purpose. However, conventional models often struggle to adequately characterize complex sea states involving mixed wind and swell systems, which exhibit multimodality and irregular dependence structures. To address this limitation, this study applies the use of Gaussian mixture models (GMM) to construct environmental contours. The GMM-based approach models the joint distribution of environmental variables in a flexible and data-adaptive manner, with the number of mixture components determined by the Bayesian Information Criterion and model parameters estimated via the expectation-maximization algorithm. Compared with the conventional conditional Weibull–Lognormal model, the GMM significantly improves fitting accuracy: the RMSE decreases from approximately 0.06 to below 0.0013, and the R² increases to nearly 1.000 across all three datasets. The KS and χ² tests confirm that the GMM adequately fits the observed data at the 0.05 significance level, whereas the baseline model is rejected in several cases. For the 100-year return period, the GMM yields maximum significant wave heights of 4.19–4.55 m with associated peak periods of 18.8–20.3 s, while the baseline model gives 4.02–4.18 m and 14.3–14.6 s, respectively. These quantitative improvements demonstrate that the mixture-based contours capture the intricate characteristics of wind–swell coexisting sea conditions more accurately, leading to enhanced representativeness of extreme sea states. Consequently, the adopted method enables more refined and reliable design sea state assessments for tested datasets, contributing to the optimization of environmental parameter selection for floating wind turbines. Full article

This study aimed to model the growth trajectories of Denizli chickens under different production systems and to identify the most appropriate nonlinear growth function within a Bayesian framework. A total of 156 birds were monitored weekly from hatch to 26 weeks of age under conventional cage, conventional floor, and enriched floor systems. Eight candidate nonlinear growth models were evaluated using Bayesian model comparison criteria, including leave-one-out cross-validation (LOO) and the widely applicable information criterion (WAIC). Among the evaluated models, the Gompertz function showed the best predictive performance, with the lowest LOOIC (225.16) and superior predictive accuracy across fit statistics. The selected model was subsequently extended to a Bayesian nonlinear mixed modelling framework to evaluate the effects of sex and production system on growth dynamics while accounting for between-animal variability. Males exhibited substantially higher asymptotic weights than females, whereas females showed faster early growth and earlier stabilization. Birds reared under the conventional floor system, particularly males, exhibited the highest asymptotic growth potential and later inflection ages, indicating a more prolonged growth phase. In contrast, enriched systems appeared to have promoted greater variability in growth responses, possibly due to increased behavioral activity and energy expenditure. The findings demonstrated that production system and sex jointly influenced both the scale and timing of growth in Denizli chickens. Beyond statistical model comparison, the Bayesian nonlinear mixed modelling approach provided biologically meaningful information that could support breeding, housing, and management decisions for indigenous and dual-purpose poultry production systems. Full article

The Poisson exponentially weighted moving average (PEWMA) control chart rests upon the equidispersion assumption of the pure Poisson distribution, a structural symmetry condition stipulating that the process mean and variance are equal. In manufacturing environments characterized by latent process heterogeneity, this assumption is systematically violated: the resulting distributions are inherently asymmetric, heavily right-skewed, and overdispersed. This structural asymmetry renders standard PEWMA control limits artificially narrow, inducing a substantial inflation of false alarm rates. This paper introduces the Poisson mixture EWMA (PM-EWMA) control chart, which models the latent heterogeneous structure of count data as a finite Poisson mixture distribution, with parameters estimated via the Expectation–Maximization (EM) algorithm without requiring prior labeling of process states. The optimal number of components is determined via the Bayesian Information Criterion (BIC) as the primary criterion, supplemented by the Akaike Information Criterion (AIC), its bias-corrected variant (AICc), and the log-likelihood ratio diagnostic. The PM-EWMA chart incorporates the exact mixture variance, accounting for both within-component and between-component variability, into the EWMA control limit structure, thereby providing a theoretically justified correction under the fitted Poisson mixture assumption. A Monte Carlo simulation study comprising 495 factorial configurations benchmarks the PM-EWMA chart against both the standard PEWMA chart and the negative binomial EWMA (NB-EWMA) chart with oracle dispersion calibration, confirming stable in-control ARL performance and demonstrating improved discrimination relative to the misspecified PEWMA baseline. Empirical validation using fabric defect count data from two textile manufacturers in Türkiye, with Overdispersion Indices of 6.01 and 2.74, respectively, demonstrates false alarm reductions ranging from 40.9% to 89.2% relative to the standard PEWMA chart, depending on the smoothing parameter and degree of overdispersion. Full article

Background: Diarrheal disease remains a major and persistent cause of morbidity and mortality in Rwanda, with substantial seasonal surges that strain routine services; however, transparent and operationally interpretable national forecasting has been underused for age-stratified burden. Methods: We analyzed the Rwanda Health Management Information System (HMIS) monthly diarrhea case counts (January 2015–December 2025), stratified by age group (under-five and five-and-above), and developed validated Seasonal Autoregressive Integrated Moving Average (SARIMA) forecasts for January 2026–December 2027. Stationarity was assessed using the Augmented Dickey–Fuller test and addressed through differencing. Candidate models were selected via rolling 5-fold cross-validation: Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), and Mean Absolute Percentage Error (MAPE) and confirmed via Ljung–Box residual diagnostics, and benchmarked against seasonal naïve, Exponential Smoothing State-Space (ETS), and Seasonal-Trend decomposition using Loess (STL) + drift reference models. Results: Rwanda recorded 6,309,098 diarrhea cases during 2015–2025, with 49.2% among under-fives; while absolute counts were higher in those aged ≥5 years, risk remained consistently higher in under-fives (91.7–229.5 per 1000) than in those ≥5 years (17.9–34.3 per 1000). Both series showed strong annual seasonality with recurrent peaks in August–November, and forecasts suggest this pattern will persist through 2026–2027. Conclusions: These findings suggest a provisional seasonal (pre-peak, peak, and post-peak) preparedness framework and age-differentiated planning signals, underscoring that burden and risk are not inter changeable across age groups. Full article

Background: Adult survivors of childhood cancer face a significant risk for treatment-related late effects that may impair health-related quality of life (HRQoL). Incorporating longitudinal changes in patient-reported symptoms beyond treatment-based risk factors may enhance the prediction of HRQoL. Methods: Survivors (n = 576) dually enrolled in the St. Jude Lifetime Cohort Study and Childhood Cancer Survivor Study reported 37 symptoms across 10 domains at three time points over 20 years to ascertain longitudinal symptom change patterns. HRQoL was subsequently assessed using SF-36 scores. Prediction models were developed using Bayesian Information Criterion Elastic Net (BIEN), first including demographic, diagnosis, and treatment variables, then adding symptom change patterns. Prediction of suboptimal HRQoL (score < 40) was evaluated using 10-fold cross-validated area under the receiver operating characteristic curve values (AUC). Results: Participants (median baseline age 26.7 years, 52% female, 90% non-Hispanic white, 41% leukemia, and 30% Hodgkin/non-Hodgkin lymphoma survivors) most frequently reported symptom domains of sensory, pain, and anxiety (50–60% at any time point), followed by depression and memory (40–50%). Consistent absence throughout follow-up was the most common pattern (41.7–98.1%), while patterns requiring symptom presence at ≥1 time point were less common (0.0–16.7%), and persistent presence throughout follow-up was rare (0.0–6.8%). Across 10 SF36-HRQoL scores, symptom-enhanced models improved prediction over non-symptom models (AUCs 0.75–0.85 vs. 0.56–0.66; p-values < 0.001). Conclusions: Longitudinal symptom change patterns substantially improved future HRQoL prediction, achieving prediction accuracy that may be of clinical effectiveness. This supports regular symptom assessment and further research towards symptom-informed risk stratification in survivorship care. Full article