Medical residency is a demanding training stage characterized by high levels of stress and burnout. As digital natives, current medical trainees (ie, residents) are frequent users of social media; however, little is known about how their personal (nonprofessional) use relates to burnout and social media addiction (SMA). This study aims to characterize the prevalence of SMA among Chinese medical trainees and explore its complex relationships with social media use patterns, occupational burnout, and related risk and protective factors. A nationwide cross-sectional survey was deployed through Wenjuanxing and disseminated via WeChat between August 29 and September 10, 2024. Data included demographics, physical and psychiatric health history, work variables (eg, training year and night shifts), personality traits, and social media use. SMA was assessed using the Bergen Social Media Addiction Scale. Logistic regression was performed to identify predictors of addiction, and mediation and moderation analyses were conducted to clarify the role of occupational burnout. Of 3621 medical trainees, 211 (5.8%) met the criteria for SMA (Bergen Social Media Addiction Scale ≥24, indicating addiction). Second-year medical trainees reported the highest addiction prevalence (92/1159, 7.9%). Logistic regression analysis revealed that higher burnout (odds ratio [OR] 1.41, 95% CI 1.23-1.62; P<.001), longer daily use (OR 1.39, 95% CI 1.23-1.56; P<.001), physical health problems (OR 1.56, 95% CI 1.13-2.16; P=.006), and psychiatric history (OR 2.00, 95% CI 1.41-2.84; P<.001) significantly increased the odds of addiction, whereas conscientiousness was protective (OR 0.92, 95% CI 0.86-0.99; P=.02). Social media use showed significant U-shaped associations with burnout, physical health problems, psychiatric history, personality characteristics, and mental health outcomes. For example, medical trainees using social media 1 hour or less (104/404, 25.7% with psychiatric history) and more than 4 hours daily (97/419, 23.2% with psychiatric history) both had higher risk profiles than moderate users. Mediation analysis showed that occupational burnout explained 28.1% of the effect of psychiatric history and 29.6% of the effect of physical health problems on addiction risk. This large-scale survey provides the first systematic characterization of SMA among Chinese medical trainees and elucidates its associated risks and protective factors. Burnout consistently emerged as a key and pervasive predictor of SMA, functioning both as an independent risk factor and as a mediator amplifying the impact of health-related vulnerabilities. Moreover, the findings highlight that both minimal and excessive daily social media use may signal distinct behavioral manifestations of distress, potentially reflecting different clinical phenotypes: digital disengagement under acute stress versus compulsive engagement driven by chronic burnout. Notably, while mental health symptoms exhibited U-shaped associations with usage, SMA risk increased progressively with daily duration. These results underscore the need for interventions that extend beyond simply monitoring usage duration, emphasizing strategies to reduce burnout and enhance the overall well-being of medical trainees.
TikTok pregnancy-related information content has not yet been investigated. To assess the quality, reliability, and misinformation on TikTok videos regarding induction of labor (IOL). A cross-sectional analysis of TikTok videos, employing the "Induction of Labor" keyword, was conducted on the 13th of January 2025. All videos retrieved under this search term were evaluated. The TikTok materials were compared between patients and healthcare with the following tools: Patient Education Materials Assessment Tool (PEMAT A/V), the modified Development of a Quality Index for Health Information (mDISCERN), global quality scale (GQS), and video information and quality index (VIQI). One hundred fifty TikTok videos were examined. The contents were created mainly from patients 52% (78/150), 39% from healthcare (59/150), and 9% (13/150) from other sources. Healthcare content showed a higher PEMAT A/V for actionability and understandability median score, 81.8% and 66.7%, respectively, compared to the patient-generated content median score of 75.0%, and 33.3% (P = 0.01 and P < 0.001). On VIQI, healthcare videos outperformed patients' content, in information accuracy (4.0 vs 2.5), precision (4.0 vs 2.5), and total VIQI score (14.0 vs. 10.0; all P < 0.001). Healthcare and other sources had a median of 2.0 for mDISCERN reliability (P < 0.001). GQS showed a median of 4.0 for healthcare content versus 2.5 median for patients' content (P < 0.001). Patients' TikTok content reporting low scores on all validated assessment tools. Healthcare videos reported a higher score of understandability, actionability, and accuracy. These findings suggest that obstetric healthcare content on social media are probably necessary to offer IOL evidence-based information.
Problematic internet use (PIU) and problematic social-media use have been associated with depressive symptoms and suicidal behaviors among university students, with limited Mediterranean evidence. This study examined their associations with stressful life events, depressive symptoms, and suicidal behaviors. A cross-sectional anonymous online survey conducted among undergraduates at the Cyprus University of Technology. Participants completed Internet Addiction Test-20 (IAT-20) to assess PIU risk, Bergen Social Media Addiction Scale (BSMAS) to assess problematic social-media use, Center for Epidemiologic Studies Depression Scale (CES-D) to assess depressive symptoms, Life Events Scale for Students (LESS-36) to assess stressful life events, and Suicidal Behaviors Questionnaire-Revised (SBQ-R) to assess suicidal behaviors. Correlation and multivariable linear regression analyses examined associations with depressive symptoms and suicidal behaviors. 1002 students completed the survey (45% response rate); 67.7% were female. PIU risk was minimal (51.1%), mild (38.6%), and moderate (10.3%). BSMAS and LESS-36 scores correlated with depressive symptoms (ρ = 0.47; ρ = 0.30) and suicidal behaviors (ρ = 0.24; ρ = 0.31; all p < 0.001). Adjusted analyses showed depressive symptoms were associated with female gender, mild-moderate PIU, problematic social-media use, and stressful life events. Suicidal behaviors were associated with male gender, non-Cypriot nationality, family history of mental illness, screen time, mild-moderate PIU, stressful life events, and depressive symptoms. Problematic internet and social-media use and stressful life events were associated with depressive symptoms and suicidal behaviors; longitudinal research is needed to clarify temporal relationships.
This study aimed to develop strategies to prevent accelerated drug release (alcohol-induced dose dumping) from modified-release matrix tablets in hydroethanolic media. Drugs with different solubility profiles (theophylline, propranolol HCl, paracetamol, and carbamazepine) were formulated with water-soluble or water-insoluble matrix formers, with optional addition of soluble or insoluble fillers. Drug release, medium uptake, and leaching were evaluated in 0.1 N HCl containing 0, 20, or 40% (v/v) ethanol. No release acceleration was observed for theophylline and propranolol HCl, which showed low solubility ratios between hydroethanolic and aqueous media (approximately 2). In contrast, paracetamol and carbamazepine (solubility ratio approximately 20) required formulation adjustments. For paracetamol, comparable release profiles across media were achieved using matrix formers with low medium uptake (e.g., Klucel® MXF or Kollidon® SR) or by incorporating a soluble filler such as lactose. For carbamazepine, the increased solubility in hydroethanolic media shifted the release mechanism from erosion-dominated to diffusion-dominated; similar release profiles were only obtained with hydrophilic polymers exhibiting relatively high erosion rates.
Menstruation has long been framed primarily as a hygiene issue, with mainstream products and public messaging emphasizing concealment and disposal of menstrual blood (MB). This has contributed to a culture of silence in which conversations about menstrual health have been marginalized in public and clinical settings. Recent international guidance, including the World Health Organization's call to reframe menstruation as a health issue, underscores the need for more open discourse. Simultaneously, social media has become a prominent space where menstruating individuals share experiences, seek advice, and challenge stigma. The resurgence of reusable menstrual products has increased users' direct observation of MB, prompting questions about variations in color, texture, and smell. These developments highlight growing curiosity about MB yet reveal persistent information gaps regarding how MB is understood outside the clinical setting. This study aimed to examine how MB is represented in social media discourse and to explore individuals' perceptions of MB's potential use as a diagnostic tool. We conducted a cross-sectional, convergent mixed methods social listening study combining qualitative content analysis, social network analysis, sentiment analysis, and descriptive statistical analysis. Data were collected from TikTok (ByteDance), Facebook (Meta), Instagram (Meta), and Reddit using Mention and Apify. Between February 1 and 28, 2025, 6263 posts and videos were extracted using 3 strategies-group searches, hashtag searches, and social listening alerts. All data were anonymized, and demographic information was unavailable. After removing duplicates, non-English content, images, and posts without reference to blood, 349 posts were included. Coding followed a multistep deductive process in Atlas.ti. All posts were assigned with quotations, which were designated with one or more codes. Network analysis examined associations between appearance descriptors and reported health conditions. Sentiment analysis assessed perceptions of MB-based diagnostics. Among the included posts (n=349), most originated from Reddit and Facebook. Seeking help (154/349, 44.1%) was the most common type of post. Appearance descriptions (n=243 posts) focused on color, particularly brown, bright red, pink, and black; consistency, particularly coagulation; and smell, mainly unpleasant. Network analysis linked specific colors and textures to perceived conditions, including miscarriage, endometriosis, hormonal changes, polycystic ovary syndrome (PCOS), and infections. Discussion of MB as a diagnostic tool (n=80 posts) was less frequent but included predominantly positive quotations (110/115, 95.7%), emphasizing accessibility, noninvasiveness, and home-based sampling. Concerns (19/115, 16.5%) focused on inclusivity, stigma, and bodily autonomy. This study demonstrates that social media serves as an important source for discussion on MB-related topics and highlights a gap between public information needs and the available scientific evidence. The findings also indicate a strong interest in MB characteristics and support further research into its diagnostic potential. To our knowledge, this is the first study to analyze social media discussions on MB characteristics and its diagnostic potential.
Social media has transformed how academics disseminate research, but its effect on academic job outcomes remains unclear. Previous research has shown correlations between social media exposure and metrics like citation counts, but these relationships may be confounded by unobserved factors such as researcher quality or access to professional networks. We examine whether social media promotion causally affects job market outcomes in economics through a field experiment on Twitter (now X). We first collect tweets about job market papers from 519 candidates and post them from a dedicated account. We then randomize half of the posts to be quote-tweeted by established economists in the candidates' fields, and measure the effects on both online visibility and hiring outcomes. We find that posts in the treatment group receive 441% more views and 303% more likes than those in the control group. Candidates whose posts were assigned to be quote-tweeted receive one additional flyout invitation compared to the control group average of 5.4 flyouts. Furthermore, women in the treatment group receive 0.9 more job offers than women in the control group, who receive 3 offers on average. Exploring mechanisms, we find that academic reputation drives these results, with stronger effects for quote-tweets from highly cited scholars and for candidates from top institutions. Our findings suggest social media promotion causally increases research visibility and improves academic job market outcomes.
Adverse events (AEs) detected in tweets and in the FDA Adverse Event Reporting System (FAERS) provide valuable insights into patient experiences with oral hypoglycemic agents including sodium-glucose transporter 2 (SGLT2) and dipeptidyl peptidase-4 (DPP4) inhibitors. This study compared the side effects identified from tweets with those in the FAERS to identify the AEs associated with each drug. We collected AE data through tweet and the FAERS during 2017-2021. Relevant sentences in the tweets were annotated and manually labeled to identify AE terms. The data obtained from both sources were categorized according to the System Organ Class (SOC) of the Medical Dictionary for Regulatory Activities. Renal and urinary disorders were defined as the index comparator with a value of 1.0. The relative frequency of a side effect compared with the index comparator was obtained. Both drugs showed similarities in high-frequency SOCs. The largest difference between the two datasets for DPP4 inhibitors was observed for the Cardiac disorders category. It ranked 12th in FAERS but 1st in tweets data, showing a marked difference in index values (FAERS 0.75, Tweet 10.80). For the SGLT2 inhibitor, the most evident difference was in the Surgical and medical procedures category. In this category, the index from FAERS was 0.26, while that from tweets was 3.49, ranking 12th and 1st, respectively. Despite the differences in the quantity and types of side effects between the two sources, we were able to identify which clinically significant side effects patients were concerned about and worried about. People share their experiences with medicines on social media. In this study, we used Natural Language Processing to read tweets about oral antidiabetic drugs such as sodium‐glucose transporter 2 inhibitors and dipeptidyl peptidase‐4 inhibitors and found side effects that were not listed in official reporting systems. These findings suggest that social media might help identify potential safety issues.
Scattering fundamentally limits the propagation of light in complex media, yet controlling it is essential for transformative advances in imaging, sensing, and optical communication. While decades of research have established powerful methods for linear wavefront shaping, the control of nonlinear scattering remains dominated by feedback-based optimization and neural networks - approaches that lack interpretability and theoretical bounds. Here, we establish the analytic inverse theory of nonlinear wavefront shaping under open-geometry scattering conditions with circular complex Gaussian statistics. By formulating an explicit scattering tensor model, we reveal how the optimal input field emerges from the dominant eigenchannel of the tensor's spectral diagonalization. This framework directly leads to a closed-form enhancement bound for second-harmonic generation. We experimentally confirm the theory by shaping wavefronts to realize single-point focusing, multi-point focusing, and global second-harmonic signal enhancement in nonlinear scattering media. By bridging nonlinear optics and inverse wavefront control, this work transforms nonlinear wavefront shaping from an optimization-driven practice into a principled, interpretable, and prediction-capable discipline.
Interactions between microplastics (MPs) and soil colloids critically influence pollutant mobility in subsurface environments, yet remain insufficiently characterized under transient flow conditions. This study explored the effects of ionic strength, ion type, and cation exchange on the co-transport and release of polystyrene (PS) MPs and soil colloids in saturated quartz sand. In NaCl solutions, pre-deposited PS particles or soil colloids increased surface roughness and chemical heterogeneity of the sand, inhibiting subsequent particle transport. PS particles preferentially attached to pre-deposited soil colloids, whereas most subsequent soil colloids adhered directly onto sand surfaces. Flushing with deionized water induced significant remobilization of both particles due to reduced ionic strength, which was well captured by a transport model coupled with ion dynamics. In CaCl2 solutions, the impact of pre-deposited particles varies with ionic strength. At lower ionic strength, competition for attachment sites promoted particle transport, while at higher ionic strength, increased surface roughness and heterogeneity inhibited mobility. Unlike Na+, Ca2+ facilitated a staggered arrangement of soil colloids and pre-deposited PS particles through cation bridging, resulting in strong retention and minimal release during deionized water flushing. Subsequent cation exchange from Ca2+ to Na+ weakened adhesive forces and triggered substantial release, with soil colloids showing greater sensitivity than PS. Overall, the interactions between suspended soil colloids and PS MPs are governed by cation valence and concentration, reflecting a balance between steric hindrance and surface heterogeneity. These findings provide insights into how pre-deposited and suspended PS and soil colloids influence each other's transport behavior and highlight the critical role of cation exchange in regulating mobility in subsurface environments.
The case of mental health disorders has been a main topic in the clinical and psychological field. The advancement of computing studies, especially in Natural Language Processing (NLP)-a subset of Machine Learning, created a system of detection that can detect the mental health state of a person in early stage to prevent the eventuality of the worst case. This is crucial since there has been a lot of case of mental health disorder-such as depression and suicide, remains undetected and untreated-especially when the internet usage is more prevalent than ever even among the most vulnerable users, which are the preadolescent users. This study explores the models that can accurately predict mental health disorder with the provided six labels the model can predict. The labels are anxiety, depression, personality disorder, stress, bipolar, and normal. The dataset is gathered from a Kaggle repository which is then processed and refined further for the training process. From multiple evaluations across diverse amount of texts from different users, our Bi-LSTM-XGBoost model outperforms the other models with an accuracy of 0.9035 and 0.4320 loss, while other models fall short within 50-84% accuracy. Further improvement can be made with our model, whether from improving the model's parameters further or by improving the quantity and quality of the dataset gathered.
The oxygen evolution reaction (OER) is a sluggish half-reaction that makes water electrolysis much less efficient. To achieve a more sustainable approach to O2 production, it is necessary to develop highly effective and stable electrocatalyst materials. In this work, a ternary hydrogen-bonded organic framework (HOF)-templated Ru-modified borophene (HOF-BB-Ru) electrocatalyst is synthesized that involves the construction of an HOF-borophene heterostructure and subsequent integration of Ru species. This ternary structure increases the accessibility of the active sites, enhancing charge-transfer properties and improving structural stability. The XRD, SEM, TEM, and XPS analyses revealed that the ternary framework was successfully formed and chemically integrated. Electrochemical analysis, including linear sweep voltammetry (LSV), chronoamperometry (CA), and electrochemical impedance spectroscopy (EIS), reveals significantly improved OER kinetics for HOF-BB-Ru. The HOF-BB-Ru catalyst exhibited low overpotential of 349 mV in an alkaline electrolyte, 420 mV in alkaline seawater (simulated), and 473 mV in natural seawater at the current density of 50 mA cm-2. These findings concluded that the HOF-BB-Ru electrocatalyst has the capability to work effectively across a wide range of electrolytic conditions, demonstrating its strength and flexibility. Finally, this study introduces a heterostructure that exhibits significant efficiency in the use of OER application in real marine water environments.
暂无摘要（点击查看详情）
Examples of a new class of synthetically straightforward and hydrolytically inert photoswitchable acylhydrazones, named photovermellogens, have been successfully developed and comprehensively studied using state-of-the-art spectroscopic and computational techniques. These compounds present photoswitchable basicity, with irradiation promoting E→Z photoisomerizations, which result in the generation of a less acidic species. The observed ΔpKa values > 1.5 within the biologically relevant window underscore the potential of photovermellogens as versatile platforms for light-triggered pH control.
Thymidine-dependent small-colony variants (TD-SCVs) are slow-growing subpopulations of bacteria that require thymidine for growth. As they do not grow on Mueller-Hinton media, conducting antimicrobial susceptibility tests on these strains is difficult, increasing the risk of overlooking antimicrobial-resistant strains. In this study, we aimed to evaluate the optimal thymidine concentration for their growth and viable cell counts on various media for thymidine-dependent small-colony variants of Enterobacterales as a preliminary investigation toward development of their antimicrobial susceptibility testing. For 11 TD-SCVs of Enterobacterales, the optimum developmental thymidine concentrations and the viable cell counts in 0.5 McFarland turbidity standard prepared from colonies cultured on various media were determined. Subsequently, we assessed a modified antimicrobial susceptibility test method. In this method, broth microdilutions were supplemented with the optimal thymidine concentration and bacterial suspensions were prepared from the colonies grown on the medium that yielded the highest viable cell count. The tested strains showed the best growth at 10 μg/mL of thymidine. The 0.5 McFarland turbidity standard prepared from colonies on Mueller-Hinton agar supplemented with 10 μg/mL thymidine had the highest number of viable bacteria. Using the modified antimicrobial susceptibility testing method with 10 μg/mL thymidine generally produced reasonable results, including ESBL-producing strains. Therefore, it is important to consider the optimal thymidine concentration for growth and the difference of the viable cell counts on each media to develop antimicrobial susceptibility testing methods for TD-SCVs.
Mobile health (mHealth) apps improve healthcare providers' accurate disease classification in resource-limited settings. Ethiopia recently introduced the National Health Data Dictionary (NHDD) mobile app for disease classification; however, healthcare providers' awareness of it remains unknown. This study aimed to assess awareness of mobile-based disease classification apps among healthcare providers working in public health facilities in northwest Ethiopia and to determine the factors associated with this awareness. A facility-based cross-sectional study was conducted among 423 healthcare providers working at 19 public health facilities in northwest Ethiopia from October 1 to 25, 2023. Data were collected using a pre-tested self-administered questionnaire. Awareness was defined as being aware of the existence of a mobile app (NHDD) for disease classification. Multilevel logistic regression analysis was used to account for clustering at the health facility level. Adjusted Odds Ratio (AOR) with 95% confidence interval (CI) was used to identify associated factors. Only 30.73% (95% CI: 26.30%-35.55%) of healthcare providers were aware of a mobile-based app for disease classification. Healthcare providers having social media accounts (AOR = 13.96; 95% CI: 2.33-83.64), ever visited the medical field by a mobile phone (AOR = 2.39; 95% CI: 1.03-5.51), digital literacy (AOR = 6.13; 95% CI: 1.50-25.01), awareness of the ESV-ICD-11 booklet on paper (AOR = 2.34; 95% CI: 1.06-5.18), and access to ESV-ICD-11 training or mentorship (AOR = 2.93; 95% CI: 1.25-6.87) were factors associated with awareness. About one-third of healthcare providers are aware of the mobile-based disease classification app. Social media use, digital literacy, prior mobile use for the medical field, familiarity with the paper-based ESV-ICD-11 booklet, and ESV-ICD-11 training or mentorship were associated factors with awareness. Targeted awareness creation interventions could be considered to support the success of mobile-based app implementation in Ethiopia.
Aging, trauma, genetic predisposition, and lifestyle impact the human body´s cartilage degradation. Once injured, continuous insults resulting from daily activities can trigger pathological conditions such as osteoarthritis, the primary cause of disability and socioeconomic loss worldwide. The main goal of all orthopedic surgeons treating joint cartilage injuries has been to reduce the extent of the lesion anatomically, repair the cartilage surface, and reestablish joint stability. Animal models are essential for the development of therapeutic drugs, but current models for cartilage defects are unsatisfactory. Osteochondral lesions in sheep are a valuable model for testing new therapies and biomaterials that can aid in the recovery of cartilage and bone in human joint environments. This study established an efficient protocol for inducing acute osteochondral defects in large animals. A standardized lesion was created in both medial femoral condyles. One knee was randomly assigned to receive gelatin-methacryloyl treatment, while the contralateral knee served as a control. Six months after the surgical procedure, the femoral condyle area was removed, the dissected knee joints were decalcified, embedded in paraffin, and cut into sections, which were stained with hematoxylin and eosin. Scores were used to evaluate the lesion. This methodology allows immediate macroscopic observation after injury induction. Additionally, this model effectively replicates clinical cartilage defects, providing a valuable model for studying their pathology and developing innovative therapeutic approaches.
Recent advances in quantum communication and quantum error correction (QEC) have motivated hybrid architectures that exploit quantum resources to enhance multimedia transmission. However, practical quantum hardware remains constrained in qubit count, making it unrealistic to apply full scale QEC to every pixel of an image. To address this, we propose a hybrid framework combining Adaptive Multi-Qubit Encoding (AMQE) with selective Quantum Low-Density Parity-Check (QLDPC) protection. The proposed solution is demonstrated for the case of image data. Our work provides a resource-efficient pathway for high-quality quantum media transmission. The system partitions the image into blocks and assigns an importance score based on local variance. High-importance blocks structural features are encoded into multi-qubit superposition states and embedded into the logical subspace of a high-rate lifted-product QLDPC code. Low-importance blocks background are transmitted with lightweight AMQE encoding. We model the channel using realistic amplitude - damping noise. Numerical simulations show that this selective protection strategy decouples perceptual quality from physical noise limits. The proposed architecture maintains a Peak Signal-to-Noise Ratio (PSNR) above 40 dB in noise regimes where classical baselines fail. The framework also retains high-structural fidelity, maintaining the Structural Similarity Index Measure (SSIM) more than 0.98, confirming robust preservation of key visual features under amplitude - damping noise. Furthermore, we demonstrate that the proposed QLDPC architecture outperforms Quantum Polar codes at finite block lengths due to the steeper error suppression slope of the Belief Propagation - Ordered Statistics Decoding (BP - OSD).
Since the rise of freely accessible pornographic streaming websites, pornography consumption has become widespread and normative worldwide. In Flanders, early exposure-before age 13-has tripled over the past decade, and frequent use, particularly among young men, is common. While pornography consumption may support body satisfaction, self-exploration, and self-esteem, evidence on its effects on sexual development and sexual well-being remains limited. Public debates are polarized, swinging between moral panic and denial of potential risks. Care providers and helplines increasingly report young people struggling with pornography-related concerns, such as self-perceived porn-induced sexual dysfunctions. Adolescents and young adults from diverse backgrounds express a clear need for guidance in navigating sexually explicit media, particularly when communication with parents, teachers, or health care providers is difficult. This project aims to generate evidence-based insights into the complex relationships between pornography consumption, sexual development, and sexual well-being among young people. By producing actionable knowledge, it seeks to inform education, prevention, and care practices that help adolescents and young adults navigate sexually explicit media in ways that promote healthy and inclusive sexual well-being within Flanders' ethnically and sexually diverse society. The project consists of four interconnected work packages: (1) examining pornography in relation to societal norms and inequalities, (2) exploring pornography within family-based sexual development, (3) investigating pornography's role in health care contexts, and (4) developing evidence-based pornography literacy tools for education and prevention. A mixed methods approach will combine systematic scoping reviews, a nationally representative survey, laboratory studies, qualitative interviews and focus groups, and co-creation with key societal stakeholders. The project received funding from Research Foundation - Flanders in 2024, and researchers were appointed between September and November 2024. Scoping reviews began in January 2025 and concluded in October 2025. A large-scale survey will be conducted between January and March 2026, followed by subsequent stages of analysis, dissemination, and valorization, concluding in 2028. Although empirical results are not yet available, the project will deliver new evidence on how pornography consumption shapes sexual development and sexual well-being across diverse contexts. It will produce practical outputs for education, health care, and policy, and contribute to reducing stigma and misinformation around pornography use. By addressing pornography as a multifaceted social and sexual phenomenon, this multidisciplinary research will advance scientific understanding and promote more inclusive, evidence-based approaches to sexual health education, care, and policy.
This study maps and analyses how tobacco companies leverage American Chamber of Commerce (AmCham) chapters globally to advance commercial interests and obstruct public health policies. We conducted a descriptive, cross-sectional document analysis using systematic keyword searches of AmCham and US Chamber websites, publications and social media, supplemented by tobacco industry monitoring platforms and media reports, to identify country-level instances of tobacco-company membership, leadership roles and tobacco-related policy interventions. Of 195 countries searched, 103 had AmCham chapters; 80 of these had tobacco-company members, around a quarter had tobacco executives in leadership or advisory roles, and almost 60% had at least one documented activity aligned with tobacco-industry positions. While opposition to tobacco taxation and price measures remained a prominent theme, a substantial and growing share of documented activities involved the promotion of industry-framed 'harm reduction' or 'smoke-free' narratives, alongside the amplification of corporate social responsibility (CSR) and environmental, social and governance (ESG) initiatives. These actions undermine implementation of the World Helath Organization Framework Convention on Tobacco Control (WHO FCTC) Article 5.3, which mandates protection of public health policies from tobacco industry interference. This study builds on earlier work by expanding and updating the landscape of AmCham involvement and providing a systematic global mapping that brings together both historical and more recent evidence on AmCham involvement in tobacco-related policy debates, offering governments and advocates an evidence-based foundation to respond to industry influence, using tools like the Global Resource Database and the List of Industry Actors, which is operated by the WHO FCTC Knowledge Hub (KH) and Global Center for Good Governance in Tobacco Control (GGTC), in compliance with the mandate provided by the tobacco control treaty's governing body, to document and monitor tobacco industry engagement in AmCham networks.
The widespread use of social media has facilitated the recognition of personality from user-generated online content. While numerous applications exist across diverse domains, such as recommender systems, most current studies focus on superficial, statistical, and explicit user content, thereby neglecting latent knowledge. In this study, we propose a method for uncovering latent psycholinguistic understanding at deeper levels of user data to enhance personality prediction through natural language processing. The proposed approach leverages fine-tuning of a domain-specific Bidirectional Encoder Representations from Transformers (BERT) model for sentence-level feature extraction and enriches the output by incorporating emotional information. This process emphasizes salient words through a single-way attention mechanism. Our single-way attention mechanism propagates information from highlighted words to the overall extracted knowledge. Subsequently, using the embeddings from the previous stage as node features, we construct a graph. A dynamic, task-oriented learning approach is then employed to determine the graph edges, using a neural network to connect different pairs of nodes. Finally, a graph neural network is combined with a classifier to predict personality traits. Experimental results demonstrate the effectiveness of the proposed model, achieving 80.27% accuracy on the Essays dataset and outperforming existing approaches. Furthermore, several ablation studies were conducted to investigate the impact of various components and parameters of the proposed architecture.