Search Results (647)

Background/Objectives: Alzheimer’s disease (AD) is an irreversible neurodegenerative disorder that imposes a profound burden on global public health. While resting-state functional magnetic resonance imaging (rs-fMRI)-based dynamic functional connectivity (dFC) analysis has demonstrated promise in capturing time-varying brain network abnormalities, existing deep learning methods suffer from three fundamental limitations: (1) an inability to model temporal dependencies across dynamic connectivity windows, (2) reliance on post hoc black-box explainability tools, and (3) misalignment between feature learning and classification objectives. Methods: To address these challenges, we propose MambaKAN, an end-to-end interpretable framework integrating a Variational Autoencoder (VAE), a Selective State Space Model (Mamba), and a Kolmogorov–Arnold Network (KAN). The VAE encodes each dFC snapshot into a compact latent representation, preserving nonlinear connectivity patterns. The Mamba encoder captures long-range temporal dynamics across the sequence of latent representations via input-selective state transitions. The KAN classifier provides intrinsic interpretability through learnable B-spline activation functions, enabling direct visualization of how latent features influence diagnostic decisions without post-hoc approximation. The entire pipeline is trained end-to-end with a joint loss function that aligns feature learning with classification. Results: Evaluated on the Alzheimer’s Disease Neuroimaging Initiative (ADNI) dataset across five classification tasks (CN vs. AD, CN vs. EMCI, EMCI vs. LMCI, LMCI vs. AD, and four-class), MambaKAN achieves accuracies of 95.1%, 89.8%, 84.0%, 86.7%, and 70.5%, respectively, outperforming strong baselines including LSTM, Transformer, and MLP-based variants. Conclusions: Comprehensive ablation studies confirm the indispensable contribution of each module, and the three-layer interpretability analysis reveals key temporal patterns and brain regions associated with AD progression. Full article

Background/Objectives: The diagnosis of endometriosis (EM) remains challenging due to the lack of a perfect diagnostic standard and the poor concordance between clinical symptoms and lesion severity. Although laparoscopy is widely used in clinical practice, it is invasive and associated with a non-negligible false-negative rate, while serum CA125 has limited diagnostic accuracy. In our previous studies, circulating endometrial cells (CECs) were identified in the peripheral blood of patients with EM, suggesting their potential as a non-invasive biomarker. Building on this finding, the present study aimed to systematically evaluate the clinical value of CECs in the diagnosis and stratified diagnosis of EM in the absence of a perfect diagnostic reference standard. Methods: Female patients treated at the Department of Obstetrics and Gynecology, Peking University People’s Hospital, between June 2022 and June 2024 were enrolled. Participants were clinically classified according to laparoscopic evaluation into an EM group and a non-EM group. However, laparoscopy was not treated as a definitive diagnostic gold standard in the statistical analysis. Instead, given the absence of a perfect reference standard, nonparametric latent class analysis was applied to jointly estimate disease status and the diagnostic performance of CECs, CA125, and laparoscopy. Patients with EM were further stratified according to dysmenorrhea severity (mild, moderate, and severe), lesion activity status (active or dormant), and menstrual cycle phase. Peripheral blood samples were collected from all participants, and CECs were detected using subtraction enrichment combined with immunofluorescence and fluorescence in situ hybridization (SE-iFISH). Serum CA125 levels were measured concurrently. Results: A total of 302 participants were included. The primary analysis focused on 133 surgically confirmed EM patients and 146 non-EM controls. After adjustment for an imperfect diagnostic reference standard, CECs demonstrated superior diagnostic performance compared with serum CA125 in the overall cohort, with higher sensitivity (0.58 vs. 0.37) and specificity (0.81 vs. 0.75). Under laparoscopic assessment in patients with severe dysmenorrhea (VAS ≥ 7), where the sensitivity and specificity were 0.759 and 1.00, respectively, CECs demonstrated superior diagnostic performance compared with serum CA125, with higher sensitivity (0.694 vs. 0.355) and specificity (0.946 vs. 0.429). Similarly, in patients with active EM, where laparoscopy showed a sensitivity of 0.79 and a specificity of 1.00, CECs again demonstrated superior diagnostic performance compared with CA125 (sensitivity 0.73 vs. 0.35; specificity 0.96 vs. 0.31), showing high concordance with laparoscopic diagnosis. When stratified by menstrual cycle phase, CECs maintained superior diagnostic performance over CA125 during both the proliferative and menstrual phases, with higher sensitivity (0.84 vs. 0.44) and specificity (0.83 vs. 0.65). Conclusions: Circulating endometrial cells (CECs) demonstrate high diagnostic accuracy for EM, significantly outperforming serum CA125, and show high concordance with laparoscopic diagnosis across clinically relevant stratified conditions in the absence of a perfect diagnostic gold standard. Full article

The low participation of men in primary education is a persistent and structural phenomenon that cannot be adequately understood through homogeneous views of teachers’ motivations and experiences. This study is conducted in the Chilean context, which is characterized by a highly feminized teaching workforce and persistent challenges related to working conditions, social valuation of teaching, and teacher retention. It aims to analyze profiles of male primary school teachers, considering their motivations, perceptions, and the meanings they attribute to the teaching profession. A sequential explanatory mixed-methods design (QUAN → qual) was employed. First, 144 male in-service primary teachers completed the FIT-Choice scale and a latent class analysis was conducted. Subsequently, in-depth interviews were carried out with an intentionally selected subsample of 20 teachers, which were analyzed using qualitative content analysis. Three distinct motivational profiles were identified: altruistic, pragmatic, and critical. The qualitative findings complemented these profiles, highlighting the influence of personal trajectories and working conditions on teachers’ career choice and retention in the profession. Overall, the findings suggest that policies for training, support, and professional induction must recognize teacher heterogeneity and promote inclusive working environments, moving beyond approaches that focus exclusively on increasing the number of men in primary education. Implications for the design of policies aimed at attracting and retaining male primary school teachers are discussed. Full article

Understanding Alzheimer’s Disease (AD) progression using resting-state functional Magnetic Resonance Imaging (rs-fMRI) remains an open challenge. Variational Autoencoders (VAEs) provide compact representations of high-dimensional neuroimaging data but lack mechanisms to highlight disease-relevant features. We propose KS-VAE, a novel framework that integrates the Kolmogorov–Smirnov test into the latent space of VAEs to identify statistically significant variables discriminating healthy from pathological brain states. This integration enables measurement of latent space shifts associated with cognitive decline, offering a quantitative approach to neurodegenerative processes. By modifying the most relevant variables, KS-VAE generates synthetic samples that simulate transitions between clinical conditions while preserving anatomical plausibility. The method enhances the modeling of temporal and distributional dynamics underlying disease progression and provides interpretable analysis of class-relevant features. Applied to rs-fMRI scans of 220 subjects from the ADNI cohort, KS-VAE demonstrated robust class separation between cognitively normal and Alzheimer’s disease subjects, achieving a classification accuracy of 84.5% and an F1-score of 84.5%, and clinically consistent synthetic transitions. KS-VAE thus offers a statistically grounded and clinically interpretable framework for understanding Alzheimer’s disease progression. Full article

Background: To investigate the effects of probiotics and their postbiotics on mouse health, this study utilized healthy mice randomly assigned to a control group (CK, n = 6), a probiotic group (L, n = 6, oral gavage 200 μL Pediococcus lactis), and a postbiotic group (PL, n = 6, oral gavage 200 μL Pediococcus lactis postbiotic). Methods: Following 21 days of continuous intervention, changes in gut metabolic profiles, microbial community structure, tissue morphology, and tight junction protein expression were systematically analyzed using metabolomics, 16S rRNA sequencing, hematoxylin and eosin (HE) staining, and immunohistochemistry techniques. Results: The results revealed that screening for significantly altered endogenous metabolites identified core differences concentrated in metabolites related to intestinal barrier repair, anti-inflammation, and antioxidant activity (e.g., 3-indolepropionic acid, astaxanthin, hydroxybenzoic acid). 16S rRNA sequencing revealed that the overall community structure was relatively stable according to principal component analysis, although differences were detected in specific taxa. However, LEfSe analysis identified significantly enriched functional microbial groups at multiple taxonomic levels in the PL group: phylum: Actinomycetota; class: Coriobacteriia; order: Coriobacteriales, Erysipelotrichales; family: Erysipelotrichaceae, Eggerthellaceae; genus: norank_Erysipelotrichaceae, Intestinimonas. These results suggest that although the overall community structure remained relatively stable, specific taxa may have differed between groups. Hematoxylin and eosin staining revealed no pathological lesions in intestinal tissues from either group, with intact mucosal architecture. Immunohistochemistry demonstrated significantly elevated expression of intestinal tight junction proteins Claudin 1, MUC-2, Occludin, and ZO-1 in the PL group compared to the CK group (p < 0.001). Conclusions: In summary, this probiotic (Pediococcus lactis) and its postbiotic showed promising effects, which may be related to changes in specific microbiota taxa, intestinal metabolic profiles, and tight junction protein expression. Beyond maintaining gut microbiota and tissue homeostasis, it enhances intestinal barrier function, suppresses latent inflammation, and boosts antioxidant capacity. Postbiotics may exhibit superior efficacy compared to probiotics. This provides robust experimental evidence for its development and application in gut health products for healthy populations. However, these findings still require further validation in studies with longer intervention periods and in disease models. Full article

Background/Objectives: Patients with breast cancer show substantial heterogeneity in terms of psychological adjustment following diagnosis. We aimed to characterize longitudinal trajectories of quality of life (QoL) and depressive symptoms during the first 18 months post-diagnosis and to identify robust clinical, psychosocial, and behavioral predictors associated with distinct adjustment pathways. Methods: Women (N = 538; mean age 55.4 years; range 40–70) with operable breast cancer (stages I–III) were drawn from the multicenter BOUNCE cohort. QoL (Global Health Status/QoL scale of the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30) and depressive symptoms (depression subscale of the Hospital Anxiety and Depression Scale) were assessed at baseline and months 3, 6, 9, 12, 15 and 18. Latent class growth analysis and growth mixture modeling identified distinct trajectory classes. Associations between early predictors and trajectory membership were examined using logistic regression combined with elastic net regularization. Results: Depression trajectories demonstrated heterogeneity, with groups characterized by persistent resilience (59.7%), stable moderate/high (25.3%), delayed onset (5.0%), and recovery (10.0%). QoL trajectories ranged from stable excellent (13.2%) and stable high (40.7%) to moderate (31.4%) and persistent low/deteriorating (6.9%), as well as a distinct recovering trajectory (7.8%). Trajectory differentiation was primarily driven by psychological resources, symptom burden, functional status, and coping processes, alongside specific contributions from clinical factors. Conclusions: Distinct subgroups of women with breast cancer follow divergent adjustment pathways. These findings highlight the multidimensional nature of resilience and support the need for tailored interventions that promote long-term well-being beyond simple risk reduction. Full article

The large-scale deployment of autonomous vehicles (AVs) in mixed-traffic environments raises an important question: how do human drivers evaluate safety when interacting with AVs under real-world uncertainty? This study aims to examine how drivers’ objective knowledge of AVs shapes their perceived safety when sharing the road with AVs in mixed-traffic environments. Using survey data from 905 licensed drivers in Wuhan, China, this study treats perceived road-sharing safety as an interaction-level evaluative outcome rather than merely a precursor of adoption intention. Latent class analysis was first used to identify knowledge-based driver segments, structural equation modeling was then applied to estimate Theory of Planned Behavior (TPB)-related psychological constructs, and ordered logit regression was finally employed to examine the determinants of perceived safety across segments. The results indicate that behavioral intention consistently shows a positive association with perceived safety; however, attitude toward AVs exhibits a significant negative association among high-knowledge drivers. This attitudinal reversal challenges the implicit homogeneity assumption embedded in conventional TPB applications and suggests that cognitive familiarity may recalibrate, rather than amplify, technological optimism. Overall, the findings show that knowledge-based heterogeneity changes the psychological mechanisms underlying safety appraisal in mixed traffic. These insights carry important implications for differentiated communication strategies and trust calibration in transitional automated mobility systems. Full article

Accurate identification of road network intersections is essential for urban planning, autonomous navigation, and traffic safety analysis. However, standard approaches relying on local geometric attributes often overlook essential topological information. This limitation is particularly problematic for intersection types that are locally similar but topologically distinct. To address this, we propose a hybrid framework that augments intrinsic node attributes with Generalized Random Dot Product Graph embeddings and neighbor-aggregated features. We utilize tree-based ensemble classifiers, specifically Random Forest and Extreme Gradient Boosting, to process this enriched feature set. Unlike standard spectral methods that assume homophily, this approach explicitly models heterophilous connectivity to capture structural patterns where dissimilar nodes connect. Experiments on a real-world urban road network demonstrate that this topological augmentation yields consistent and robust improvements. The proposed integration with the Extreme Gradient Boosting model achieves a Macro ROC AUC of 0.8966 and a Micro F1 score of 0.7005, outperforming the baseline model (ROC AUC 0.8100, Micro F1 0.5919). Performance gains are most pronounced for topologically ambiguous intersection classes, confirming that local attributes alone fail to capture structural distinctions. These results demonstrate that latent structural context is a critical discriminator for granular road intersection classification. Full article

Background/Objectives: Complementary medicines (CMs) are widely used in Australia, yet consumer beliefs about their safety and effectiveness often diverge from the scientific evidence. Contemporary statewide data from Victoria, particularly about these perceptions and underlying perception profiles, are limited. We therefore aimed to characterise CM use patterns and perceptions of it among Victorian adults and identify the demographic and use-related belief patterns. Methods: A cross-sectional survey was conducted in metropolitan and regional Victoria (November 2024–August 2025) among adults (≥18 years) who had used complementary medicines in the previous 12 months (N = 447). The questionnaire assessed CM use patterns, perceived effectiveness, safety, quality, perceived risk relative to prescription medicines, adverse events, and demographics. The analyses included descriptive statistics, χ² tests with multiple-comparison control, Spearman correlations, and a multivariable regression. An exploratory factor analysis (EFA) and latent class analysis (LCA) were used to identify the perception dimensions and distinct consumer profiles. Results: CM use was frequent (62.2% daily; 19.2% weekly) and often long term (>1 year, 55.0%). The most commonly used products were vitamin D (53.0%), multivitamins (39.8%), magnesium (34.5%), iron (33.8%), and vitamin C (30.0%). The perceptions were favourable: 77.3% rated CMs as effective, 90.4% as safe, and 60.3% as high quality; 78.5% perceived CMs to have lower side-effect risks than prescription medicines. Adverse events were reported by 12.3%. In the adjusted models, adults ≥ 65 years and monthly/occasional users were less likely to endorse “lower risk than prescription medicines” (aOR: 0.18; 95% CI: 0.06–0.51; aOR: 0.36, 0.18–0.72). East Asian respondents had lower odds of endorsing CM effectiveness than Caucasian/White respondents (aOR: 0.28, 0.11–0.72). Their perceived quality was higher among men (aOR: 1.73, 1.09–2.74) and adults aged 55–65 years (aOR: 3.81, 1.39–10.48). Conclusions: In this contemporary statewide Victorian sample, CM use was common and generally viewed positively, yet the comparative risk may be underestimated. Profiling perception patterns and identifying belief patterns by age, culture, and use intensity provides actionable targets for clinician/pharmacist counselling and culturally tailored education to support safer, evidence-aligned CM use. Full article

Chryseobacterium indologenes MA9 is a causative agent of root rot disease in Panax notoginseng (P. notoginseng), with its high incidence being a major manifestation of continuous cropping barriers, severely hindering the sustainable development of the P. notoginseng industry. In this study, a novel lytic bacteriophage, MA9V-3, was isolated from wastewater, targeting C. indologenes MA9. The phage produced clear plaques, ranging from 1 to 3 mm in diameter, with a surrounding halo. Phage MA9V-3 achieved an adsorption rate of up to 80% after 30 min of contact with C. indologenes MA9, a latent period of approximately 40 min, and an average burst-size if 160 PFU/cell. Transmission electron microscopy revealed that phage MA9V-3 possesses an icosahedral head and a contractile tail, exhibiting a typical myovirus-like morphology. According to the latest ICTV taxonomy, MA9V-3 belongs to the class Caudoviricetes, and the phage’s biocontrol efficacy and inhibitory capacity were evaluated at different multiplicity of infection (MOI s). The results showed that the highest titer recorded at 1.6 × 10¹⁰ PFU/mL. Whole-genome sequencing revealed that MA9V-3 is a double-stranded circular DNA virus, with a genome length of 103,203 bp, GC content of 34.29%, and 150 open reading frames (ORFs), one of which is related to tRNA. Only 13 of these ORFs encode known functional sequences, likely due to the limited available gene data for such phages in the database, with additional details on hypothetical proteins yet to be uncovered. Comparative database analysis confirmed that the phage genome contains no antibiotic resistance or toxin-related genes. Phage therapy experiments were performed using MA9V-3 and two other phages screened in our laboratory. The experimental results showed that phage MA9V-3 may be a potential candidate for effectively controlling the infection of Panax notoginseng by C. indologenes MA9, and offering valuable insights into the potential application of phage therapy for managing bacterial plant diseases. Full article

This study evaluates the ability of three classification procedures to distinguish areas with different levels of atmospheric pollution, based on biomonitoring carried out by analyzing the color and spectral characteristics of mulberry (Morus L.) and linden (Tilia L.) leaves. Sampling was carried out in areas that were grouped into four classes according to the concentrations of fine particulate matter (PM2.5, PM10) and gaseous pollutants (TVOC, NOx, SOx, CO, and eCO₂), measured using a specialized multisensor device. A total of 57 informative features were analyzed, representing indices obtained from two color models (RGB and Lab), as well as from VIS and NIR spectral characteristics measured for the adaxial and abaxial leaf surfaces. The data processing methodology includes feature selection using the ReliefF method and a comparative analysis between two approaches to dimensionality reduction—principal components (PC) and latent variables (LV). The results indicate that data reduction using PC provides significantly higher accuracy and better class separability, regardless of the classifier used, compared to LV, where errors exceed 40%. The comparison between classifiers shows a clear superiority of nonlinear models. While linear discriminant analysis demonstrates low efficiency, quadratic discriminant analysis (Q and DQ) and SVM with radial basis function (RBF) achieve high accuracy of class separability, reaching 100% in the SVM-RBF model for both tree species. The study also reveals functional asymmetry: the adaxial side of the leaves is more informative for spectral indices, while the abaxial side is more sensitive to color changes. The results confirm that the combined optical characteristics obtained from the leaf surface of bioindicators form a reliable method for ecological monitoring of air quality in urban areas. Full article

Global progress towards the 2030 Sustainable Development Goals (SDGs) remains critically off track, with current trends indicating that only 17% of targets will be met by the deadline. As sustainability transitions increasingly depend on regional and institutional capacity, understanding heterogeneous transition pathways and resilience across territorial contexts is essential. This study investigates whether observed divergence in SDG performance reflects temporary setbacks or persistent structural regimes characterised by distinct institutional and technological configurations. Using panel data from over 160 countries (2019–2024), we employ annual latent class analysis to identify hidden structures in SDG performance across 15 goals, introducing intertemporal volatility as a dimension of development dynamics. We complement this with ordered logistic regression to examine structural determinants of regime membership, including governance quality, digital infrastructure, health investment, and macroeconomic indicators. Our analysis identifies three temporally stable development regimes—lagging, transitional, and leading—with fewer than 15% of countries transitioning between classes over the observation period. ANOVA results reveal that internet access and government effectiveness exhibit the most substantial between-regime differences. Ordered logit models indicate that governance quality and digital connectivity are the strongest correlates of regime membership (government effectiveness: β = 0.943, p < 0.001; internet penetration: β = 0.049, p < 0.001), whereas short-term GDP growth exerts negligible influence (p > 0.10). These findings challenge assumptions of linear convergence in sustainable development and provide a data-driven framework for evaluating transition dynamics across diverse territorial contexts. The results suggest that achieving the SDGs requires that deep structural constraints be addressed—particularly digital divides and institutional quality—through regionally targeted policy design rather than relying solely on incremental adjustments or economic growth. The identified regimes provide a basis for place-based targeting by distinguishing contexts where governance and digital capacity constraints are binding. Full article

Background/Objectives: Sleep is an important biobehavioral process that supports child and adolescent health and development. However, many prior studies examining sleep and mental health have relied on total sleep scores, which may mask the heterogeneity of sleep disturbances. Youth exposed to childhood adversity are at increased risk for sleep disturbances and poor mental health, and thus it is important to examine the links between sleep and mental health within adversity-exposed samples, such as those involved with the child welfare system. Methods: This study used latent class analysis to identify underlying patterns of sleep disturbances and examine differences in mental health symptoms (assessed at baseline and at an 18-month follow-up) across the identified subgroups in a sample of child welfare-involved adolescents (N = 1041, M_age = 13.63 years, SD = 1.86). Our sample was derived from the second cohort of the National Survey on Child and Adolescent Well-Being (NSCAW) study. Results: We identified three subgroups of sleep disturbances: no sleep disturbances (38%), sleeping more than peers and overtired (16%), and trouble maintaining sleep (47%). We found significant mean differences in mental health symptoms across subgroups. Across internalizing, externalizing, and post-traumatic stress disorder (PTSD) symptoms at baseline and at an 18-month follow-up, those in the no sleep disturbances subgroup had overall lower levels of symptoms compared to those in the trouble maintaining sleep subgroup, which had higher levels of symptoms. Compared to those in the sleeping more than peers and overtired subgroup, the trouble maintaining sleep subgroup had higher levels of PTSD symptoms at baseline, and higher levels of externalizing and PTSD symptoms at the follow-up. Those in the sleeping more than peers and overtired subgroup had significantly higher levels of internalizing, externalizing, and PTSD symptoms at baseline compared to the no sleep disturbances subgroup, but there were no significant differences at the 18-month follow-up. Conclusions: The current study highlights the importance of considering the heterogeneity of sleep disturbances to identify child welfare-involved youth who may be more at risk for sleep disturbances and poor mental health and to inform more targeted sleep interventions for this population. Full article

Deep learning models for brain tumor classification from magnetic resonance imaging (MRI) often achieve high in-dataset accuracy but exhibit substantial performance degradation when evaluated on unseen clinical data due to domain shift arising from variations in imaging protocols and intensity distributions. Existing approaches largely rely on architectural scaling or parameter-level regularization, which do not explicitly constrain the stability of learned feature representations. This manuscript proposes Feature Space Stability Regularization (FSSR), a lightweight and model-agnostic training framework that enforces consistency in latent feature representations under realistic, MRI-safe-intensity perturbations. FSSR introduces an auxiliary feature space loss that minimizes the ${\ell }_2$ distance between normalized embeddings extracted from the input MRI images and their intensity-perturbed counterparts, alongside standard cross-entropy supervision. This manuscript evaluated FSSR across three convolutional backbones, ResNet-18, ResNet-34, and DenseNet-121, trained exclusively on the Kaggle Brain MRI dataset. Feature space analysis demonstrates that FSSR consistently reduces mean feature deviation and variance across architectures, indicating more stable internal representations. Generalization is assessed via zero-shot evaluation on the fully unseen BRISC-2025 dataset without retraining or fine-tuning. On the source domain, the best-performing configuration achieves 97.71% accuracy and 97.55% macro-F1. Under domain shift, FSSR improves external accuracy by up to 8.20 percentage points and the macro-F1 by up to 12.50 percentage points, with DenseNet-121 achieving a 96.70% accuracy and 96.87% macro-F1 at a domain gap of only 0.94%. Confusion matrix analysis further reveals the reduced class confusion and more stable recall across challenging tumor categories, demonstrating that feature-level stability is a key factor for robust brain MRI classification under domain shift. Full article

Open Educational Resources (OER) have emerged as a cost-effective alternative to traditional commercial textbooks in higher education, towards the goal of alleviating college students’ financial burden of educational expenses. However, mixed findings about the influences of the integration of OER on student learning are present. To address the gap, this study investigated whether student motivation in OER served as a latent factor that impacts their academic achievement in online asynchronous courses offered in public universities. Particularly, this study (N = 247) implemented an advanced person-centered approach—stepwise latent class analysis—to profile student achievement goals in an OER-based course and examined their relationships with academic achievement. The 7-point Likert responses were collapsed into three categories to address sparse response distributions. The analysis identified four latent classes based on students’ responses to a validated survey aligned with the 2 × 2 achievement goal theory framework, including highly ambitious, cautious, strategic, and low-goal learners. Subsequent analysis revealed that these four latent classes showed differences in academic achievement as well as task value and expectancy beliefs. The implications of these results for researchers and college instructors and future research directions are discussed. Full article