PLOS ONE Alerts: New Articles

Effect of nutrition counseling on complementary feeding practices among mothers with children aged 6–23 months in West Gojjam Zone, Northwest Ethiopia: A cluster-randomized controlled trial

2026-10-29T14:00:00Z

by Shiferaw Birhanu, Getu Degu Alene, Yeshalem Mulugeta Demilew

Background

Appropriate complementary feeding is essential for children’s growth and cognitive development. However, these practices remain inadequate in the study area. Therefore, this study aimed to evaluate the effect of nutrition counseling on complementary feeding practices among mothers with children aged 6–23 months in West Gojjam Zone, Northwest Ethiopia.

Methods

A two-arm, parallel, cluster-randomized controlled trial was conducted from June 2023 to December 2024 in rural districts of West Gojjam Zone. Thirty clusters (15 per arm) were assigned to either an intervention group (n = 401), which received Health Belief Model-based nutrition counseling by Women’s Development Army leaders, or to a control group (n = 401), which received routine education. The intervention was delivered monthly for nine months from infant age 6–8 months. The primary outcome was appropriate complementary feeding practices. Secondary outcomes included minimum dietary diversity, meal frequency, acceptable diet, and egg/flesh food consumption. Data were collected via interviewer-administered questionnaires. The intervention effect was analyzed using Generalized Estimating Equations with adjustment for potential confounders. Statistical significance was set at p < 0.05.

Results

The difference-in-differences in appropriate complementary feeding practices between groups was 13.2 percentage points. After adjustment, mothers in the intervention group were nearly 2.5 times more likely to practice appropriate complementary feeding than controls [AOR = 2.49, 95% CI: (1.40, 4.42)]. Children of counseled mothers were also more likely to achieve minimum dietary diversity [AOR = 1.75, 95% CI: (1.04, 2.93)], minimum meal frequency [AOR = 1.62, 95% CI: (1.01, 2.61)], and minimum acceptable diet [AOR = 2.13, 95% CI: (1.26, 3.59)] compared with controls.

Conclusion

Health Belief Model-based nutrition counseling was positively associated with improvements in appropriate complementary feeding practices, minimum dietary diversity, meal frequency, and acceptable diet. Integrating such counseling into routine maternal and child health services may strengthen child nutrition programs in similar rural settings.

Trial registration

ClinicalTrials.gov (NCT05871346).

Interpretable EOG-based gaze classification using tree-based models: A SHAP and LIME analysis

2026-10-29T14:00:00Z

by Ayse Nur Ay Gul

Electrooculography (EOG) is a low-cost biosignal modality for gaze-based human–computer interaction, but reliable validation and model interpretability remain important challenges for machine learning-based gaze classification. This study proposes an explainable EOG gaze classification framework using UNIOVI-EOG dataset. Horizontal and vertical EOG recordings from nine participants were converted into 100 ms non-overlapping windows, and only stable windows with at least 90% label purity were retained. Time-domain window features were extracted and evaluated using Decision Tree, Random Forest, XGBoost, and CatBoost classifiers under subject-wise five-fold cross-validation, ensuring that participants in the validation folds were not included in training. The final dataset contained 81,481 accepted windows across nine gaze classes. All models achieved very high subject-wise performance, with Random Forest providing the strongest overall results. The high performance was consistent with the controlled fixation protocol and the strong separability of horizontal and vertical EOG amplitude patterns, as supported by simple mean-amplitude baseline and leakage diagnostic analyses. Model interpretability was assessed using SHapley Additive exPlanations (SHAP) and Local Interpretable Model-Agnostic Explanations (LIME). The explanations showed that amplitude-related features, especially channel means, medians, quantiles, and absolute mean values, were the main contributors to classification decisions. SHAP–LIME rank correlation and permutation-based faithfulness analysis provided additional quantitative support for the explanations. These findings suggest that leakage-controlled, window-level, explainable machine learning can provide accurate and physiologically interpretable EOG gaze classification in controlled experimental settings.

Factors associated with choosing diagnostic radiology as a future specialty among Palestinian medical students: a multicenter cross-sectional study

2026-10-29T14:00:00Z

by Ibrahim Ammar AbuAttallah, Jana A. Qrareya, Reem S. Adawi, Jehad Khamaysa, Mahdi Thabet Aljamal

Background

Understanding factors influencing specialty choice is essential for workforce planning and curriculum development. Recruitment into diagnostic radiology remains challenging in several regions, particularly in low- and middle-income settings. This study aimed to identify factors associated with Palestinian medical students’ consideration of diagnostic radiology as a future specialty, with particular emphasis on educational exposure, perceptions of artificial intelligence (AI), and cognitive influences.

Methods

A multi-institutional cross-sectional study was conducted to evaluate factors influencing the choice of diagnostic radiology as a future specialty among Palestinian medical students between 2 November 2025 and 5 February 2026. Data were collected using a previously validated questionnaire adapted from Almuhanna et al. Descriptive statistics were performed, followed by bivariate analyses and multivariable logistic regression to identify independent predictors of considering radiology as a future specialty. Adjusted odds ratios (OR) with 95% confidence intervals (CI) were reported.

Results

A total of 427 students participated (56.7% female; mean age 21.75 ± 2.1 years). Among 300 valid responses regarding radiology intention, 102 students (34.0%) reported considering radiology as a future specialty. In multivariable analysis, lack of exposure to radiology research (OR = 0.14, 95% CI 0.06–0.34, p < 0.001) and lack of preclinical radiology teaching (OR = 0.26, 95% CI 0.12–0.54, p < 0.001) were independently associated with a lower likelihood of considering radiology. Enjoyment of diagnostic problem-solving tasks (OR = 3.35, 95% CI 1.39–8.09, p = 0.007) and exposure to tele-radiology (OR = 3.96, 95% CI 1.72–9.12, p = 0.001) were independently associated with higher likelihood. Lifestyle, impact on patient care, and work environment were the most frequently reported motivating factors.

Conclusions

Palestinian medical students’ interest in radiology is strongly associated with modifiable educational exposures, cognitive engagement in diagnostic reasoning, and perceptions of evolving technological practice. These findings support the potential value of structured undergraduate radiology exposure, research opportunities, and mentorship initiatives to improve awareness and interest in radiology.

Reported clinical breast examination activity in Bangladesh, 2019–2025: Temporal trends, geographic variation, positivity, and reporting completeness

2026-10-29T14:00:00Z

by Sourov Das, Simran Patra, Shubham Om, Sushmita Kundu

Background

Routine clinical breast examination (CBE) services contribute to breast cancer early-detection activities in Bangladesh, but national program performance and reporting completeness are poorly characterized. We assessed reported CBE activity, geographic variation, positivity, reporting completeness, and temporal trends from 2019 through 2025.

Methods

We retrospectively analyzed routine CBE program data from 1,209 location-specific eligible facilities across Bangladesh from January 2019 through December 2025. Because individual-level identifiers were unavailable, records were analyzed as CBE events rather than unique women screened. Reported CBE examination rates used estimated populations of women aged 30–69 years, with district denominators proportionally allocated using census-based population distributions. Reporting completeness was defined as submitted facility-month reports divided by expected reports. Interrupted time-series regression assessed changes associated with March 2020 using calendar-month fixed effects and heteroskedasticity- and autocorrelation-consistent standard errors.

Results

A total of 3,193,080 reported CBE events were recorded. The reported examination rate increased from 0.3% in 2019 to 2.4% in 2023, then declined to 1.8% in 2024 and 1.1% in 2025. In 2025, division-level rates ranged from 0.7% to 1.7% and district-level rates from 0.2% to 2.8%. Overall positivity was 1.0% (31,624 positive events), reaching 1.6% in 2025. Mean annual reporting completeness rose from 12.1% in 2019 to 42.2% in 2024 and was 39.7% in 2025; no month reached 60% national completeness. Interrupted time-series analysis found no significant immediate level change (14,416 events; 95% CI −7,438–36,269; P = 0.196) or post-intervention slope change (−375 events/month; 95% CI −1,673–923; P = 0.571).

Conclusions

Reported CBE activity expanded through 2023 but subsequently declined, with marked geographic variation and persistently incomplete reporting. These findings describe reported program activity rather than true population screening coverage. Strengthening routine reporting, individual-level tracking, quality assurance, and referral follow-up is needed for more reliable program evaluation.

Explainable machine learning framework with Granger-based temporal predictive validation for monitoring fiscal risk in national health insurance expenditure

2026-10-29T14:00:00Z

by Gwangsu Lee

Rising health insurance expenditure (HIE) poses fiscal risks and challenges to the sustainability of public insurance systems. Despite growing interest in data-driven analysis, empirical evidence on the associations between macroeconomic, social, and financial conditions and national HIE remains limited, particularly across provider types. This study introduces an explainable machine-learning framework with temporal predictive validation to analyze and forecast HIE using 60 macro-financial and socioeconomic indicators. The framework is applied to three monthly expenditure series from Korea’s National Health Insurance system (2000–2023). Predictive performance and interpretability are assessed using two univariate autoregressive baselines, three regularized linear indicator-based benchmarks, and six nonlinear indicator-based ensemble models. LightGBM serves as the primary explanatory model, and SHapley Additive exPlanations (SHAP) quantifies indicator contributions and captures autoregressive persistence. Beyond associational insights, temporal predictive validation using lagged regression and Granger-type temporal predictive tests examines, within a prespecified exploratory lag window, whether SHAP-identified indicators show statistically supported exploratory lagged monitoring relevance. These results are interpreted as hypothesis-generating temporal monitoring evidence rather than as evidence of structural causal effects. Although persistence-based baselines achieved the strongest short-term forecasting accuracy, indicator-based models provided model-conditional explanatory information beyond autoregressive controls. The results indicate that shared macroeconomic pressures and provider-specific factors are associated with HIE predictions. The cost-of-living index is consistently associated with expenditure patterns across categories, whereas domestic demand and external conditions indicate provider-specific associations for medical institutions and pharmacies. Only a subset of predictors retained exploratory lagged monitoring support, underscoring the distinction between model-based predictive importance and hypothesis-generating monitoring signals. This integrated framework may inform transparent fiscal risk monitoring hypotheses and contextual review of public HIE.

Field study on compaction effect and energy evolution characteristics of highway subgrade under low-amplitude vibration

2026-10-29T14:00:00Z

by Xiaobo Liu, Ping Xie, Wubin Wang

Subgrade compaction is a critical procedure in highway construction. To elucidate the dynamic characteristics and energy transfer mechanisms throughout the entire compaction process, this study utilized four test lanes to conduct field experiments on highway subgrade under varying numbers of rolling passes. Comprehensive measurements and analyses were performed regarding the dynamic characteristics of vibration waves, energy distribution, and the evolution of soil compactness during operation. Furthermore, the compaction efficiency and energy transfer within the subgrade were quantitatively evaluated. The results indicate that the dynamic behavior of subgrade materials varies significantly across different compaction stages, which are consequently categorized into three distinct phases: loose, optimal compaction, and over-compaction. In this experiment, the third rolling pass represented the optimal compaction state, characterized by distinct resonance peaks, a regular frequency spectrum, and highly efficient energy transfer. As compactness increases, the horizontal propagation of vibration waves strengthens, whereas vertical attenuation accelerates, causing energy to concentrate beneath the vibration source. Additionally, the proposed method for evaluating compaction efficiency based on the energy transfer coefficient corroborates the three-stage evolution of the process, offering a direct, quantitative indicator for real-time quality assessment and endpoint determination. These findings provide new theoretical insights and empirical data to support intelligent and refined process control in subgrade compaction.

Changes in ware–decoration associations in ceramic assemblages from Wacheqsa, Chavín de Huántar, Peru

2026-10-29T14:00:00Z

by Christian Mesía-Montenegro

Ceramic differentiation can involve altered associations among continuing attributes without wholesale categorical replacement. At Wacheqsa, Chavín de Huántar, Peru, this study compares ware composition, decorative techniques, and their associations in 1,308 decorated or surface-treated sherd observations from five archaeological contexts. Early Platforms and Water Flood constitute the earlier Urabarroid aggregate; Stone Rooms, Midden, and Late Platforms constitute the later Janabarroid aggregate. These labels describe ceramic affinities within pre-existing context sets, whose stratigraphic networks are only partly connected and whose formation histories differ. On identical common support of nine wares and six techniques (n = 334/822), ware-margin Hellinger distance is 0.430 (95% fixed-count interval 0.392–0.481), technique-margin distance is 0.588 (0.551–0.634), and joint distance is 0.734 (0.714–0.793). A homogeneous-association model rejects a common ware–technique association after aggregate-specific margins are fitted (G² = 118.59; pboot = 0.0001); the result persists without Midden (G² = 90.16; pboot = 0.0001). Early Platforms differs from each later context, and stamping/impression is prominent in all three later samples, although those contexts retain distinct joint profiles. Nine of eleven ware categories occur in both aggregates, with pronounced differences in the association of polished wares with incision and stamping/impression. The contrast is consistent with differing decorated repertoires across activity and depositional settings, including contexts associated with collective consumption. Local superpositions and the published chronology support a temporal interpretation, but the contribution of elapsed time cannot be separated from differences in activity and deposition.

Gynecological malignancy signal triage: A two-dimensional framework integrating US-derived FAERS data with comparative US–China GBD trajectories

2026-10-29T14:00:00Z

by Ying Ma, Dandan Li, Jieru Duan, Zhuowen Sun, Qirong Dou, Tao Wu, Na Gao

Background

Cancer-related adverse-event reports are difficult to interpret because statistical disproportionality, stimulated reporting, and background disease burden are distinct phenomena. To address this, we formalized a two-dimensional prioritization framework that combines FAERS signal credibility with independent GBD burden context. Because the FAERS dataset is overwhelmingly of US origin, we utilized China exclusively as a prespecified macro-epidemiological GBD comparator, strictly avoiding cross-national pharmacovigilance extrapolations.

Methods

We analyzed FAERS reports (2004–2023) and female GBD 2023 estimates for China and the United States. Four disproportionality algorithms (ROR, PRR, IC, and EBGM) were evaluated for 150 high-frequency primary-suspect drug–event pairs (50 per cancer group). We report any-algorithm positivity and four-algorithm concordance alongside case-definition sensitivity analyses. GBD trends, age–period–cohort models, and exploratory projections to 2053 were integrated to provide population-level context, explicitly not as exposure denominators or for causal validation.

Results

FAERS cohorts yielded 19,717 cancer-associated report records. Among 150 high-frequency candidate pairs, 42 met all four algorithmic thresholds. The top-50 candidate screen captured approximately 72.8%–75.8% of primary-suspect records, serving as a variance-reduction step rather than a complete all-drug screen. Drugs like ranitidine and talc demonstrated marked disproportionality but also exhibited reporter and temporal features heavily compatible with stimulated reporting. Time-to-onset distributions reflected recorded administrative reporting intervals rather than biological carcinogenesis. GBD estimates revealed divergent China–US trajectories, notably a rising baseline for US uterine cancer.

Conclusions

The formalized F × G framework operates as a transparent triage tool, not a causal classifier. FAERS disproportionality generates hypotheses; GBD trajectories indicate potential public-health priority. All prioritized signals—particularly those involving consumer products or immunosuppressants—require formal causal evaluation in longitudinal exposure-denominator datasets.

Soil texture and depth differentially regulate total and available nutrient stoichiometry in Yilong Lake basin agroecosystem, China

2026-10-29T14:00:00Z

by Yongru Lai, Luoqi Zhao, Chenyang Zhang, Zijun Du, Yandan Wang, Tianyi Pu, Change Liu, Changqun Duan

The stoichiometric ratios of soil available nutrients serve as a pivotal indicator for assessing the sustainability of farmland and nutrient cycling efficiency within fragile ecosystems. Soil texture and depth are significant factors influencing nutrient availability, however, their regulatory mechanisms within highland lakeshore agricultural systems remain poorly understood. This study, conducted in the Yilong Lake watershed of the low-latitude plateau in southwest China, collected 216 soil samples. It innovatively determined both total organic carbon, nitrogen, and phosphorus (SOC, TN, TP) and available carbon, nitrogen, and phosphorus (DOC, AN, AP) simultaneously. Path analysis was employed to examine the direct and indirect associations between soil texture (Clay, Loam, Clay Loam, Sand) and depth (0–20 cm, 20–40 cm, 40–60 cm) on nutrient ratio measurements. Findings indicate: (1) Distinct nutrient limitation and leaching risk patterns emerged across soil textures: Loam exhibited strongest carbon and nitrogen sequestration capacity; Although Clay Loam exhibits reduced phosphorus availability due to iron and aluminium oxide immobilisation, it demonstrates pronounced nitrogen limitation rather than phosphorus limitation; Sand though rich in available phosphorus, had the lowest C:N, suggesting heightened carbon and nitrogen leaching risks. (2) Soil depth significantly altered nutrient ratios: topsoil (0–20 cm) showed marked nutrient enrichment, while subsoil (40–60 cm) exhibited elevated TN:TP, suggesting that nitrification may have intensified with increasing depth. (3) Path analysis revealed that texture directly regulates nutrient measurement ratios (DOC:AN) and indirectly governs C:N:P balance by influencing SOC, AP, and AN content; depth directly or indirectly affects measurement relationships through pathways such as TP. This study elucidates the key regulatory roles of soil texture and depth in nutrient ratios, particularly available nutrient balance, providing direct scientific evidence for implementing differentiated, precision nutrient management based on soil physical properties in Yilong Lake basin agroecosystems and other ecologically fragile areas.

Genomic dissection of environmental responsiveness and cross-environment productivity in fenugreek

2026-10-29T14:00:00Z

by Maha Aljabri, Mustafa M. H. Abd El-Wahab, Amani Mahmoud Dobeie, Mohamed El-Soda

Dissecting the genetic basis of genotype × environment interaction (G × E) is important for breeding crops with stable or responsive performance across production environments. Using previously published seed-yield data from 107 Egyptian fenugreek accessions evaluated under two contrasting field environments, we investigated whether yield potential and environmental responsiveness share a common genetic basis. Seed yield per plant was measured in Giza (SYP_G; reference environment) and Wadi El-Natrun (SYP_W; contrasting environment). Nine derived indices were calculated to characterize environmental responsiveness, productivity, and stability: relative performance index (RPI), yield plasticity index (YPI), stress tolerance index (STI), log response ratio (LRR), mean productivity (MP), harmonic mean (HM), tolerance index (TOL), stress susceptibility index (SSI), and AMMI stability value (ASV). The near-zero correlation between SYP_G and SYP_W (r = 0.107) indicated substantial genotype reranking across environments, while several response indices were closely related to SYP_W and therefore predominantly reflected environment-dependent yield expression. Phenotypic analyses also identified contrasting performance strategies, ranging from highly responsive accessions, such as G15 (RPI = 3.629), to highly stable accessions, such as G41 (YPI = 0.001). These phenotypes were integrated with 38,142 SNPs using univariate association-mapping approaches. In addition, SYP_G and SYP_W were subjected to multivariate association-mapping. Association mapping identified one genome-wide-significant SNP for STI and several suggestive loci showed concordant associations across mapping methods. Within the current panel, marker coverage, and significance thresholds, most adaptation-index associations did not overlap previously reported yield loci; however, dDocent_Contig_39939_66, previously associated with seed yield in this panel, also showed cross-method suggestive associations with RPI and TOL. Comparative genomic analysis of three legume reference genomes supported two gene-linked candidate loci associated with STI: a cation/H⁺ antiporter and an importin subunit beta-1. The same analysis classified two of the strongest statistical signals, including the sole genome-wide-significant locus, as repetitive or transposable-element sequences rather than single-copy genes. These findings demonstrate that statistical association strength does not necessarily establish a coding-gene basis for loci identified using reference-free ddRAD markers. Within the current population size and marker coverage, environmental responsiveness and productivity appeared partially distinct but biologically connected components of environment-dependent yield performance. The results provide a genomic and phenotypic framework for prioritizing fenugreek germplasm for further evaluation while emphasizing the need for independent-population, multi-season, and functional validation of the identified candidate loci.

Lethal and sublethal influence of triflumezopyrim on life history traits and enzyme activities in the bean bug, Riptortus pedestris (Hemiptera: Alydidae)

2026-10-29T14:00:00Z

by Gefan Tian, Jianglong Guo, Jingjie An, He Yuan, Zhixu Hu, Yaofa Li, Zhanlin Gao

Riptortus pedestris (Fab.) is an important soybean pest in East Asia, where its piercing-sucking feeding behavior leads to soybean syndrome. Triflumezopyrim, a novel mesoionic insecticide, has been widely applied to control sap-sucking insects. However, its lethal and sublethal impacts on R. pedestris remain poorly understood. This study investigated the lethal and sublethal effects of triflumezopyrim on R. pedestris using bioassays, age-stage two-sex life table analyses, and enzyme activity measurements. Results demonstrated that triflumezopyrim exhibited strong toxicity (LC₅₀ = 7.46 mg/L) and notable repellent activity against R. pedestris. Exposure of nymphs to sublethal concentrations (LC₁₀ and LC₃₀) significantly prolonged nymphal development and the adult preoviposition period, while reducing adult longevity, oviposition duration, and fecundity in the parental (F₀) generation. In addition, LC₃₀ exposure produced adverse transgenerational effects in the F₁ generation, including extended adult preoviposition period, decreased male longevity, reduced oviposition duration, fecundity, net reproductive rate (R₀), intrinsic rate of increase (r), and finite rate of increase (λ) of the population. Moreover, both LC₁₀ and LC₃₀ treatments led to substantial elevation in the activities of antioxidant (superoxide dismutase and peroxidase) and detoxification (multifunctional oxidase and glutathione S-transferase) enzymes. These findings indicate that triflumezopyrim exerts strong acute toxicity and pronounced sublethal inhibitory effects on R. pedestris, ultimately suppressing population growth potential. This study offers a scientific basis for the rational use of triflumezopyrim as well as the development of strategies for the management of this species.

Study on the association between the low-density lipoprotein to high-density lipoprotein ratio and clinicopathological characteristics of breast cancer: A retrospective analysis incorporating subgroup evaluations by age, body mass index, and menopausal status

2026-10-29T14:00:00Z

by Lei Dong, Can Liu, Lixin Li, Xiaotong Zhang, Wei Xu, Gaosai Dai

Breast cancer (BC) represents the most prevalent malignant tumor among women globally, with its pronounced heterogeneity presenting significant challenges for precision diagnosis and treatment. Aberrant lipid metabolism, particularly the ratio of low-density lipoprotein to high-density lipoprotein (LHR), has garnered considerable interest in relation to BC development and progression. Nevertheless, the association between LHR and clinicopathological characteristics remains inadequately explored. To address this gap, the present retrospective study analyzed 583 BC patients who underwent surgical treatment at Qilu Hospital of Shandong University. Preoperative blood lipid profiles and clinicopathological data were systematically collected to investigate the association between LHR and critical features, including molecular subtypes. The findings indicated that LHR levels were significantly elevated in patients aged 46 years or older, postmenopausal women, and individuals with a body mass index of 25 kg/m² or higher. Subgroup analyses incorporating age, menopausal status, and BMI suggested potential interactions between these clinical factors and LHR. Although univariate analysis demonstrated an association between LHR and certain clinicopathological features, multivariate logistic regression analysis revealed that menopausal status, rather than LHR, was the sole independent predictor of ER-positive and PR-positive status. Therefore, LHR should not be considered an independent predictive biomarker for receptor status. In conclusion, although the LHR demonstrates statistically significant associations with certain clinicopathological characteristics, its modest effect sizes and discriminative ability close to chance indicate limited clinical utility as an independent biomarker. These preliminary findings highlight the need for future research focused on identifying reliable metabolic signatures. Such studies should consider utilizing multi-marker panels instead of relying exclusively on individual lipid metrics for risk stratification in breast cancer.

Loss of cohesin deacetylation through KDAC8 inactivation leads to altered cohesin complex expression and chromosomal instability

2026-10-29T14:00:00Z

by Tasha B. Toro, Razan H. Hammad, Thomas M. Huckaba, Terry J. Watt

The cohesin complex is critical for both proper gene expression and for pairing sister chromatids during mitosis to ensure proper chromosome segregation. Acetylation and deacetylation of SMC3, a component of cohesin, is a critical regulatory mechanism controlling stability and reformation of the complex. Patients with Cornelia de Lange Syndrome have been observed to have mutations in cohesin-related genes, including in the lysine deacetylase KDAC8 (HDAC8), leading to disruption of complex formation and recycling. To further understand the cellular consequences of disrupting the SMC3 acetylation cycle, we utilized cell lines in which KDAC8 is replaced by a catalytically-inactive variant (8CDm) or knocked-out (8KO). Although the 8CDm cell line completes mitosis at the same rate as wild-type cells, 8CDm cells display severe chromosomal instability, characterized by increased presence of micronuclei and a significant subpopulation of cycling tetraploid cells. These defects were not observed in 8KO cells. Although total SMC3 protein levels are similar in all cell lines, SMC3 and other cohesin-associated genes are differentially expressed in the mutant cell lines. We hypothesize a model in which 8KO cells compensate for the inability to deacetylate SMC3 by increasing both SMC3 turnover and gene expression, and 8CDm mediated suppression of SMC3 turnover and expression results in fewer functional cohesin complexes and the severe chromosomal instability phenotype. Overall, these observations provide a greater understanding of the importance of SMC3 deacetylation by KDAC8 and insight into the phenotypic range observed in Cornelia de Lange Syndrome patients.

Expression of Concern: Pedestrian POSE estimation using multi-branched deep learning pose net

2026-10-29T14:00:00Z

by The PLOS One Editors

Correction: Molecular Mechanism of Mutant p53 Stabilization: The Role of HSP70 and MDM2

2026-10-29T14:00:00Z

by Milena Wiech, Maciej B. Olszewski, Zuzanna Tracz-Gaszewska, Bartosz Wawrzynow, Maciej Zylicz, Alicja Zylicz

Retraction: Aesthetic preference for artificially selected color variant affects mate choice copying behavior in female Poecilia latipinna

2026-10-29T14:00:00Z

by The PLOS One Editors

Assessment of light sensitivity using the Ocular Photosensitivity Analyzer: Discomfort threshold, pain threshold, and tolerance in healthy volunteers

2026-10-29T14:00:00Z

by Nathan Goturi, Cameron Talbert, Nicholas Pondelis, Mariela C. Aguilar, Alex Gonzalez, Cornelis Rowaan, Heather Durkee, Jean-Marie Parel, Vanda Faria, Eric Alan Moulton

Photophobia encompasses a continuum of light-induced experiences from mild discomfort to overt pain and intolerance, yet clinical tools rarely quantify these distinctions, leaving many patients undiagnosed and untreated. This study employed the Ocular Photosensitivity Analyzer (OPA), a compact instrument for quantitative sensory testing (QST) with light, to investigate three perceptual levels in healthy adults: visual photosensitivity discomfort threshold (VPDT), visual photosensitivity pain threshold (VPPT), and an exploratory outcome, visual photosensitivity tolerance level (VPTol). Twenty-nine participants ages 19−34 underwent standardized OPA testing, during which they identified when light first became uncomfortable, painful, and intolerable. After each trial, participants rated the peak pain intensity (PI) and peak unpleasantness (UP) using visual analog scales. Measured illuminance values differed significantly across the perceptual levels (RM-ANOVA, N = 28, F(1.69, 45.6)=37.06, p < 0.001, η_p² = 0.58; Mauchly’s W = 0.82, p = 0.072), indicating sensory levels are differentiable. Post hoc t-tests revealed VPTol exceeded VPPT (N = 28, t(27)=−2.29, p = 0.030, d_z = 0.43)), and both exceeded VPDT (VPTol: N = 29, t(28)=−7.69, p < 0.001, d_z = 1.43, VPPT: N = 29, t(27)=−7.69, p < 0.001, d_z = 1.47). Friedman’s test revealed UP (N = 28, χ²(2)=17.64, p < 0.001, Kendall’s W = 0.32) and PI (N = 28, χ²(2)=22.36, p < 0.001, Kendall’s W = 0.40) differences across perceptual levels and post hoc Wilcoxon tests showed UP and PI increased between VPDT and VPPT (UP: N = 29, Z = −4.34, p < 0.001, r = 0.57, PI: N = 29, Z = −4.58, p < 0.001, r = 0.60) and UP increased between VPDT and VPTol (N = 28, Z = −2.92, p = 0.007, r = 0.39), suggesting that both affective and sensory dimensions of pain vary across perceptual levels. Although preliminary, these findings are consistent with a multidimensional framework of photophobia. Additionally, the OPA’s ability to differentially quantify VPDT, VPPT, and VPTol offers a novel approach to characterize light sensitivity that may aid future assessment of abnormal light sensitivity.

Identification of prognostic biomarkers for nasopharyngeal carcinoma through microRNA‑mRNA network analysis combined with screening of immune‑related genes

2026-10-29T14:00:00Z

by Lihe He, Xuan Huang, Ping Huang

Background

Nasopharyngeal carcinoma (NPC) is a common head and neck cancer, and radiotherapy is its main treatment. Recurrence and metastasis are the main causes of treatment failure. Although immunotherapy can prolong patient survival, the molecular mechanisms underlying NPC development and progression remain unclear. MicroRNAs (miRNAs) and their target genes are abnormally expressed in many tumors and are closely involved in tumor development. Therefore, this study aimed to identify key miRNAs and immune-related target genes in NPC, construct miRNA–mRNA regulatory networks, and screen for and validate core immune genes to identify prognostic molecular biomarkers for NPC.

Methods

We downloaded NPC-related miRNA (GSE32960) and mRNA (GSE12452, GSE13597, GSE64634) microarray datasets from the Gene Expression Omnibus (GEO) and identified differentially expressed miRNAs (DEMs) and mRNAs (DEGs) using R. Target genes of DEMs were predicted with TargetScan, miRDB, and miRWalk, and then intersected with DEGs to identify core genes. A miRNA–mRNA regulatory network was constructed in Cytoscape. The core genes were further overlapped with immune-related gene sets, followed by functional enrichment analysis. A prognostic risk model was established using LASSO–Cox regression, and immune cell infiltration was evaluated with CIBERSORT. Model validation, immunohistochemical analysis, and drug–gene interaction analysis were also performed. In vitro, RT–qPCR was conducted in C666-1 and NP69 cell lines to verify the expression of key genes.

Results

We identified 213 potential target genes of key miRNAs. Three miRNAs (hsa-miR-30e-5p, hsa-miR-30a-5p, and hsa-miR-20a-5p) were associated with NPC prognosis, and 17 core immune-related genes were enriched in cytokine-mediated signaling, chemokine response, and JAK-STAT pathways. Five genes (CXCL11, SOCS1, ROBO1, LGR5, and RORB) showed promising diagnostic value for NPC; their expression was positively correlated (P < 0.05) with CD8 ⁺ T-cell, macrophage, and neutrophil infiltration. Drug-gene analysis predicted that 5-fluorouracil targeted LGR5 and aldesleukin targeted SOCS1. RT-qPCR demonstrated that SOCS1 was upregulated approximately 3.7-fold in NPC cells (P < 0.01).

Conclusion

This study identified potential novel key genes in nasopharyngeal carcinoma, providing new therapeutic targets for the treatment of this disease.

Integrative genomic SEM analysis of shared genetic architecture and candidate genes for insulin resistance–related metabolic traits

2026-10-29T14:00:00Z

by ShuYan Li, Yi Liang, Guangwen Zhu, Lijiao Wu, Hong Li, Yu Li, Qiu Chen

Insulin resistance is a complex metabolic state involving impaired insulin sensitivity, compensatory insulin secretion, abnormal lipid metabolism, and altered fat distribution. However, the shared genetic architecture underlying these correlated phenotypes remains incompletely defined. In this study, we applied Genomic Structural Equation Modeling to genome-wide association summary statistics of fasting insulin, insulin sensitivity index, triglycerides, and waist-to-hip ratio to construct an insulin resistance-related latent genetic factor. SNP-level association analysis was then performed for the latent factor, followed by risk locus annotation, fine-mapping, transcriptome-wide association analysis, gene-level prioritization, enrichment analysis, and exploratory structural biology assessment of selected missense substitutions. The single-factor model showed acceptable but not optimal approximate fit and captured a biologically consistent genetic pattern, with positive loadings for fasting insulin, triglycerides, and waist-to-hip ratio and a negative loading for insulin sensitivity index. The latent factor showed detectable SNP heritability and revealed 3,067 genome-wide significant candidate SNPs, with Q_SNP retained as a complementary SNP-level heterogeneity diagnostic.. Functional annotation identified 100 genomic risk loci and 1,530 FDR-significant genes in MAGMA. Fine-mapping prioritized multiple candidate regions, including loci near PPARG, RSPO3, SLC38A11, SLC39A8, BCL2, GDF5, and SPATA5. TWAS and FOCUS further highlighted candidate genes involved in adipocyte biology, glucose–lipid metabolism, insulin secretion, and expression regulation, including PPARG, MLXIPL, TCF7L2, FAM13A, CCDC92, and SLC22A3. Enrichment analyses suggested that genetic signals associated with the insulin resistance-related latent factor were mainly linked to fat distribution, cardiometabolic traits, adipose tissue, mesenchymal cell lineages, and active regulatory genomic regions. In a separate exploratory analysis, AlphaFold3 modeling, ThermoMPNN prediction, and molecular dynamics simulations suggested structural perturbations for three ultra-rare missense substitutions in nominated candidate proteins: BCL2-W188G, GDF5-C400Y, and RSPO3-C129Y. These substitutions were selected using functional-impact annotations and modeling feasibility, were not GWAS lead or fine-mapped causal variants, and do not establish the mechanisms underlying the GWAS associations. These findings provide an integrated genetic and structural framework for understanding shared mechanisms underlying insulin resistance-related metabolic abnormalities, while emphasizing the need for further experimental validation.

Few-shot recognition of lane-changing intention in connected vehicle environment: A meta-transformer approach

2026-10-29T14:00:00Z

by Xiaoqian Liu, Wenxin Jiang, Yining Ren

This paper proposes a Meta-Transformer model for few-shot accurate lane-changing intention recognition in connected vehicle environment. Given the velocity, acceleration, distance headway, time headway and time to collision of a vehicle at past time steps, a transformer model is developed to recognize the lane changing intention of the vehicle at the current time step. The multi-head attention layer of the Transformer model is capable of effectively extracting features from the multivariate time serie input. This enhances the accuracy of lane-changing intention recognition. The Transformer model is combined with a model -agnostic meta-learning (MAML) algorithm to be able to perform well with small amount of training data. The meta-learning consists of meta-training and meta-testing. During the meta-training, a task is sampled from the distribution over tasks. The tasks include recognition of intention of lane-keeping, lane-changing to left lane, and lane-changing to right lane. The model is trained with few samples and feedback from the loss function defined on the tasks. The loss function is defined as the categorical cross entropy of the errors between the labels and the results generated by the proposed model. During the meta-testing, the model is tested on new data from the distribution of the transition from a feature and the label to another feature. The model is trained by minimizing the test error. Experimental results show that the proposed model can achieve 8.7 percentage points higher accuracy than a transformer model.

Enhanced enfortumab vedotin sensitivity in a cisplatin-resistant bladder cancer model

2026-10-29T14:00:00Z

by Taisuke Tobe, Jun Teishima, Takuto Hara, Yusuke Shiraishi, Hideto Ueki, Naoto Wakita, Yasuyoshi Okamura, Yukari Bando, Kotaro Suzuki, Tomoaki Terakawa, Akihisa Yao, Koji Chiba, Hideaki Miyake

Objective

Enfortumab vedotin (EV) and cisplatin (CDDP) are central treatments for advanced urothelial carcinoma, yet the influence of prior CDDP exposure on subsequent EV efficacy remains unclear. In this report, we characterized EV-related drug responses in a CDDP-resistant bladder cancer subline.

Methods

A CDDP-resistant subline (HT1376-CR) was generated from parental bladder cancer cell (HT1376/P) through continuous exposure to CDDP. Sensitivities to CDDP, EV, and the EV payload monomethyl auristatin E (MMAE) were compared between HT1376-CR and HT1376/P. Nectin-4 expression, EV binding, and drug-resistance-related gene expression were evaluated. The effect of ABCB1 silencing on drug response was also examined. In vivo EV responses were explored using xenograft models.

Results

HT1376-CR cells showed increased sensitivity to EV in vitro and in vivo. HT1376-CR cells were more sensitive to MMAE and showed greater G2/M accumulation after MMAE exposure. In contrast, cell-surface nectin-4 expression and EV binding were reduced. ABCB1 was markedly upregulated in HT1376-CR, and ABCB1 silencing increased sensitivity to CDDP, MMAE, and EV within the resistant subline.

Conclusion

The CDDP-resistant HT1376 subline showed enhanced sensitivity to EV, and ABCB1 suppression further increased EV sensitivity within this resistant subline.

Eating behaviour and glycaemic control: A cross-sectional study in adolescents with diabetes

2026-10-29T14:00:00Z

by Nesibe Şimşekoğlu, İlknur Arslanoğlu, Banu Portakal, Selman Aktaş

Background

Achieving early glycemic control in adolescents with diabetes is associated with improved short- and long-term health outcomes. The aim of this study was to examine the relationship between eating attitudes and glycemic control, as well as the factors associated with eating attitudes and glycemic control, in adolescents with diabetes.

Methods

This descriptive cross-sectional study included 143 adolescents with diabetes, aged 10–19 years, who attended a research hospital between February and April 2024. The data were collected using a Descriptive Information Form, Eating Attitudes Test (EAT) and an information form questioning the values of some metabolic variables.

Results

The mean age of the participants was 14 ± 2.3 years and 49.7% had diabetes between 1–5 years. The mean total score of the EAT scale was 16.62 ± 8.30. The mean score of the “Dieting” subscale of the EAT scale was 10.39 ± 5.58, the mean score of the “Bulimia food preoccupation” subscale was 2.33 ± 3.03 and the mean score of the “Oral Control” subscale was 3.79 ± 3.08. There was no significant difference between the mean EAT total score and glycemic control.

Conclusions

The mean EAT-26 total score was below the commonly used cutoff score, and no significant association was found between EAT-26 scores and HbA1c. Differences in EAT-26 total and subscale scores were observed according to selected demographic and clinical characteristics, including BMI, treatment modality, and treatment adherence. These findings describe self-reported eating attitudes in the study sample and support the inclusion of eating-related assessment within the holistic care of adolescents with Type 1 DM.

Magnesium oxide tablets form rigid masses in xanthan gum-based thickeners: A physicochemical hazard revealed by a clinical case

2026-10-29T14:00:00Z

by Soichi Makino, Miyu Ooe, Kazunari Miyazawa, Shogo Kayano, Sayaka Kominato, Yoshinori Saito, Ryohei Hirota, Yuta Watanabe, Fumiko Hayano

Dysphagia, or difficulty swallowing, is a matter of serious concern among older adults. While thickening agents are essential for dysphagia management, they are not pharmacologically inert; their physicochemical interactions with oral medications remain poorly understood and can pose an iatrogenic hazard of mechanical esophageal obstruction. Following a case of acute esophageal obstruction of unknown origin, this study aimed to identify the foreign body, to assess the reproducibility of any disintegration failures across formulations and thickeners, and to examine the systemic barriers that would allow such a hazard to remain unaddressed. This foreign body, a magnesium oxide (MgO) tablet, caused obstruction in a malignant esophageal stricture. It was analyzed using ex vivo computed tomography and chemical analysis. A static immersion model was used to mimic esophageal stasis. The disintegration behaviors of four MgO formulations (n = 6 each) were compared using thickeners (1.0–3.0%) and a pharmaceutical swallowing aid jelly (SA-J). Furthermore, a cross-sectional market survey was conducted in nine regions in Japan to assess the economic barriers to the clinical adoption of SA-J as an alternative swallowing aid. The foreign body was identified as a non-disintegrated, swollen MgO mass. In vitro experiments confirmed a consistent phenomenon observed across all tested formulations: thickeners completely inhibited disintegration in every combination tested, increasing tablet weight by approximately 2.2-fold at concentrations of 2.0–3.0% and forming rigid masses. Conversely, the tablets disintegrated in SA-J in approximately 25 s. The market survey revealed that SA-J costs were approximately 14.5-fold higher per administration than thickeners (p < 0.001). Although based on a single case, these findings highlight that xanthan gum-based thickened water represents a physicochemical hazard that may remain unrecognized in current clinical practice. Choice of swallowing vehicle should be made with full awareness that xanthan gum-based thickeners can prevent the disintegration of MgO tablets.

Light availability, not soil texture, correlates with shrub community structure in mesic forest gaps

2026-10-29T14:00:00Z

by Sam W. Anderson, Katherine A. McCulloh

Woody shrubs are important components of temperate forest understories, contributing to forest structure, diversity, and succession. However, the extent to which variation in canopy openness and soil texture influence shrub occurrence, diversity, and composition remains unclear. Because woody tissues require comparatively high light availability and sandy soils may favor shrubs over trees, we hypothesized that shrub richness and cover would increase in canopy gaps and sandier microsites. To address this, we quantified soil texture, canopy gap structure, and recorded shrub abundance in twenty randomly stratified plots within sixteen different mesic forests in Wisconsin, USA. Using a mixed-effects modeling approach, we determined that shrub richness, Shannon diversity, and coverage increased significantly in more open canopy gaps, while soil sand content had no significant impact on these metrics. Canopy gaps also had a significant effect on the composition of shrub communities at the plot level. At the stand level, neither sandier nor more open-canopied forests had more diverse, dense shrub layers, though more heterogenous canopies had a weak negative correlation with shrub richness. Broadly, shrub abundance and cover increased in more open gaps, though individual species demonstrated significantly different preferences in terms of light availability and soil sand content. Our results demonstrate the importance of local light availability for shrub growth. However, species-specific responses to soil texture and light availability and the modeled importance of unmeasured site-level factors suggest that additional environmental and biotic factors are likely to impact this diverse temperate guild.

A target-domain augmentation and class-wise alignment framework for few-shot cross-condition bearing fault diagnosis

2026-10-29T14:00:00Z

by Yijia Hu, Ying Gu, Yanjie Su, Chuan Zhao

Cross-condition bearing fault diagnosis often has to adapt to a new operating condition with only a few labelled target samples. This setting creates two coupled problems. First, the sparse target support set cannot describe the target fault distribution sufficiently. Second, the same fault class may occupy different feature regions across source and target conditions, which weakens source-domain pretrained classifiers. Existing methods usually improve few-shot learning or reduce domain shift, but they rarely enrich the labelled target-class representation before class-wise source-target alignment. To address this limitation, this paper proposes a target-domain augmentation and class-wise alignment framework for few-shot cross-condition bearing fault diagnosis. The rime optimization algorithm (RIME) is first used to optimize the variational mode decomposition (VMD) parameters K and α, and fault-sensitive components are selected using envelope entropy and adjacent modal-frequency overlap. The reconstructed signal is converted into symmetric dot pattern (SDP) images to provide a noise-suppressed polar representation. The fast generative adversarial network (FastGAN) then generates auxiliary target-domain SDP samples from the limited labelled target support set, making each target fault class less sparse. On this basis, class-wise maximum mean discrepancy (CMMD) is embedded into a Swin Transformer classifier to align source and target features within each known fault category. Experiments on the Case Western Reserve University (CWRU) and Xi’an Jiaotong University–Shenyang Blower Works (XJTU-SY) datasets show that the proposed framework improves target-domain few-shot adaptation, achieving 93.33% and 97.58% accuracy under the G20 and G50 XJTU-SY target-augmentation settings, respectively.

The role of speaker accuracy and mutual exclusivity in children’s novel word learning

2026-10-29T14:00:00Z

by Shalini Banerjee, Ishanti Gangopadhyay

Children draw on multiple sources of information during word learning, including speaker testimony and lexical expectations. Yet, little is known about how these cues jointly influence children’s ability to learn new words. The present study examined the combined effect of speaker accuracy and mutual exclusivity (ME) on short-term retention of label-object associations in 4- to 7-year-old monolingual English speaking children. Children were presented novel and familiar object pairs, and were taught novel labels by either an accurate speaker (one who previously labeled familiar objects correctly) or an inaccurate speaker (one who previously labeled familiar objects incorrectly). Novel labels were introduced under two conditions: Align-with-ME (speakers labeled the novel object with novel names) and Conflict-with-ME (speakers labeled the familiar object with novel names, conflicting with ME). Results showed that children retained novel labels for novel objects from both accurate and inaccurate speakers when speaker testimony aligned with ME expectations. This suggests that children are open to learning novel words from speakers regardless of their prior accuracy and may also draw on other available cues, such as label–object co-occurrences, during learning. However, when speaker testimony violated ME expectations, children performed poorly, possibly due to difficulty disambiguating between speaker testimony and their existing lexical knowledge. Together, these findings suggest that children’s short-term retention of label–object associations is supported when available cues align, irrespective of speaker accuracy status. These associations, however, become less robust when ME and speaker cues provide conflicting information.

Dynamic analysis and spatiotemporal evolution of multi-stage knowledge dissemination under AI-enabled dual-channel “Interpersonal–Human-Machine” interaction

2026-10-29T14:00:00Z

by Yanhua Lv, Yadong Wang, Huanhuan Wang, Sida Kang

With the deep integration of digitalization and artificial intelligence technologies into enterprise knowledge management, AI-assisted learning is reshaping the channels through which knowledge is acquired within organizations. Meanwhile, knowledge dissemination is evolving from traditional interpersonal interaction toward human-machine collaborative interaction. In light of this transformation, this study develops a multi-stage 2SEIR-CA dynamic model of knowledge dissemination under an AI-enabled mechanism that incorporates dual-channel “interpersonal–human-machine” interaction. Specifically, employees are classified into five states: employees who have not yet mastered the knowledge, employees with limited knowledge mastery, employees with strong knowledge mastery, knowledge-disseminating employees, and knowledge-exited employees. On this basis, the next-generation matrix method is employed to calculate the basic reproduction number of the system and to analyze the existence and stability of equilibrium points. An optimal control model is then established to derive strategies for promoting knowledge dissemination, and a cellular automaton model is introduced to characterize its spatiotemporal evolution. The results show that the system exhibits a pronounced threshold effect: when key dissemination parameters exceed their critical levels, knowledge can continue to diffuse within the organization; otherwise, it will gradually decay. Second, the AI-enabled mechanism enhances knowledge transformation efficiency and contact intensity, thereby significantly promoting the transition of employees with strong knowledge mastery to knowledge-disseminating employees. Third, the optimal control strategy proposed in this study can optimize transformation pathways across different stages under resource constraints, thereby improving overall dissemination efficiency and maintaining the structural stability of the dissemination system. Fourth, the spatiotemporal simulations demonstrate that, under optimal regulation, knowledge dissemination can form a stable diffusion pattern with high coverage. Under the assumptions and parameter settings of the proposed model, the findings provide theoretical insights into the dynamic mechanisms underlying AI-enabled knowledge dissemination and offer model-based implications for knowledge management practices in organizations.

Prophylactic endoscopic transpapillary gallbladder stenting at fully covered self-expandable metal stent placement to prevent cholecystitis in malignant distal biliary obstruction: A multicenter prospective single-arm study protocol (PROTECT study)

2026-10-29T14:00:00Z

by Takayuki Oda, Yusuke Kurita, Masato Suzuki, Satoru Shinoda, Takamitsu Sato, Shinichi Nihei, Yu Honda, Takeshi Iizuka, Shin Yagi, Sho Hasegawa, Kensuke Kubota, Masato Yoneda

Background

Malignant distal biliary obstruction (MDBO) is commonly managed with endoscopic retrograde cholangiopancreatography (ERCP)-guided transpapillary stenting. Fully covered self-expandable metal stents (FCSEMSs) provide durable biliary drainage but may increase the risk of acute cholecystitis due to cystic duct obstruction. Prophylactic endoscopic transpapillary gallbladder stenting (ETGBS) at the time of FCSEMS placement has been proposed to reduce the risk of cholecystitis after FCSEMS placement. However, the available evidence is limited. This study aims to explore the potential clinical benefit of attempting prophylactic ETGBS by estimating the incidence of acute cholecystitis after FCSEMS placement and characterizing the technical feasibility and safety of the strategy in patients with MDBO.

Methods

This multicenter, prospective, single-arm exploratory study (PROTECT study) will be conducted at four institutions in Japan. In all eligible patients undergoing FCSEMS placement for MDBO, prophylactic ETGBS will be attempted. The primary endpoint is the incidence of acute cholecystitis within 30 days after FCSEMS placement. Secondary endpoints include the technical success rate of ETGBS, procedure-related adverse events, the incidence of acute cholecystitis within 30 days among patients with technically successful ETGBS, and the cumulative incidence of acute cholecystitis from day 31 until the end of follow-up. The target enrollment is 50 patients. The primary endpoint will be summarized as a proportion with a two-sided 90% Wilson score confidence interval. A two-sided 95% Wilson score confidence interval will also be reported as a supplementary analysis, and a supplementary exploratory two-sided one-sample score test will compare the observed proportion with the historical benchmark of 12.1%.

Discussion

This study will provide prospective data on the potential clinical utility, technical feasibility, and safety of attempting prophylactic ETGBS and may inform the design of future comparative studies.

Trial registration

Japan Registry of Clinical Trials number: jRCT1030250361. Registered on 12 September 2025.

Mathematical dynamic integration model based on game theory and optimization theory

2026-10-29T14:00:00Z

by Jing Wang, Xuanwei Huang, Shuai Li

As technology advances, the research on mathematical dynamic integration models has yielded a potent instrument for the analysis and optimization of complex systems, thereby fostering progress across multiple domains. To solve the issues of traditional dynamic integration models, such as the lack of solid mathematical theoretical support, imbalanced weight allocation among multiple participants, and insufficient dynamic adaptability, this study integrates the Shapley value method from cooperative game theory with Monte Carlo sampling techniques to construct an information evaluation model. The calculation of the marginal contributions of participants in different alliances enables the quantification of the value of multi-source information, thus resolving the issue of imbalanced weight allocation. Furthermore, a dynamic optimization model is built by integrating Nash equilibrium and Markov decision processes, utilizing state transition probabilities to capture temporal changes in strategies and achieve stable strategy optimization. Finally, a mathematical model that includes integration is designed. Experiments that compare the two models show that the average time it takes to make a decision goes up by only 0.2 seconds when the number of samples is between 25 and 150, and the amount of memory used goes up by only 1.2GB. In the InfiMM-WebMath-40B dataset, the model’s standard deviation increases from 0.020 in 25 samples to 0.030 in 150 samples, an increase of only 0.010. In the MV-MATH dataset, the standard deviation increases by only 0.011. Furthermore, the proposed model outperforms the comparative models in all metrics, reducing convergence time by 14.7%, training time by 15.3%, energy consumption by 16.3%, and iteration count by 21.5%, while maintaining a low central processing unit utilization rate. Therefore, the proposed mathematical dynamic integration model can more accurately handle uncertain information in complex systems, optimize system decisions, reveal the inherent laws governing dynamic changes in systems, and provide stronger theoretical support and methodological guidance for solving practical problems.

Vertical accuracy of ASTER, SRTM, and NASADEM digital surface models across Malawi’s diverse topography

2026-10-29T14:00:00Z

by Kennedy Sones Nazombe, Brighton A. Chunga, Charles Bakolo, Rodrick Mlewa, Godwell Nhamo

Digital Surface Models (DSMs) are widely used as important sources of topographic information for earth science and environmental applications. The accuracy of DSMs has direct impacts on several environmental calculations and processes. Therefore, it is imperative to select the most suitable DSM based on the aim and accuracy requirements of each application. However, the use of an appropriate DSM has always been a challenge. This study evaluated the vertical accuracy and performance of the freely available Shuttle Radar Topography Mission (SRTM), NASA DEM (NASADEM), and Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER). The DSM elevations were measured against high-precision national geodetic control points. Among other accuracy metrics, we computed Root Mean Square Error (RMSE) and examined the influence of elevation, slope, aspect, curvature, and land cover. At the national scale, SRTM (RMSE of 19.55 m) and NASADEM (RMSE of 19.99 m) showed comparable performance, while ASTER produced a higher RMSE of 22.13 m. However, the principal finding was that DSM accuracy varied substantially across Malawi’s diverse terrain. All the DSMs performed particularly well in the low-lying Lower Shire Valley, with RMSEs of 3.25 m (NASADEM), 4.30 m (SRTM), and 6.91 m (ASTER). Errors increased markedly in the Highlands, where RMSE reached 35.03 m for NASADEM, 33.62 m for SRTM, and 36.94 m for ASTER. In addition, accuracy decreased with increasing elevation and slope, particularly in highland regions and on slopes steeper than 30°. Aspect showed the highest relative importance in the Random Forest analysis. However, the differences in accuracy among aspect classes were comparatively modest, and greater underestimation was observed on north- and west-facing slopes. These findings demonstrate that national-scale accuracy statistics alone do not adequately describe DSM suitability across Malawi’s diverse landscapes. DSM selection should therefore consider the terrain and land-cover characteristics of the intended application area.