Browse Articles

Retraction: Hybrid deep learning and feature selection approach for autism detection from rs-fMRI data

Correction: Sex and BMI as predictors of pill residue in dysphagia: a multivariate analysis

Risk factors for decline in estimated glomerular filtration rate amongst Malawian adults living in rural Karonga: Protocol for a prospective cohort study using cystatin C- and creatinine-based eGFR

Background The global burden of chronic kidney disease (CKD) is rising, disproportionately impacting on low- and middle-income countries. In African populations, use of serum creatinine to estimate glomerular filtration rate (GFR) significantly underestimates CKD prevalence, contributing to its under-recognition as a health problem. Serum cystatin C provides more accurate estimates of Iohexol measured GFR than creatinine. Early diagnosis and treatment of CKD is essential to reduce premature morbidity and mortality. However, little is known about risk factors for CKD development and progression in Africa owing to limited longitudinal data. This study aims to determine risk factors for progressive kidney function decline among adults living in rural, northern Malawi. Methods This protocol describes a prospective study being conducted in a general population Health and Demographic Surveillance Site in rural Karonga, Malawi (2024–2025). Eligible participants are adults aged 18 years and over who participated in two population-based surveys of long-term health conditions (2013–2016 and 2021–2025), with availability of baseline measures. New household-level data is being prospectively collected on CKD risk factors, alongside blood and urine samples. Cystatin C and creatinine will be tested on individual-level paired, stored serum samples collected at three longitudinal time points. Urine will undergo dipstick urinalysis, microscopy and testing for albumin and creatinine to quantify proteinuria. The primary outcome will be sustained 25% reduction in estimated GFR (eGFR) from baseline and change in eGFR category, determined using serum cystatin C. Multivariable logistic regression will be used to determine effect size estimates of key risk factors for kidney function decline. Discussion This study will provide important data on risk factors for eGFR decline and CKD progression amongst Malawian adults. The findings will inform future research into important context-specific risk factors, and could directly inform health policies in Malawi for targeting CKD screening, prevention and treatment strategies to high-risk patient groups.

The impact of climate and vegetation on the riverside architecture of the Brazilian Amazon

Abstract This research aimed to investigate the impact of vegetation conservation on the local climate and architectural production of two riverside communities in one of the capitals of the Brazilian Amazon, one close to and the other far from the urban area. Based on the inductive-exploratory method, we collected primary climatological data in the external area and inside the dwellings in both communities (October 2022 to May 2023), developed analyses of the Normalized Difference Vegetation Index (NDVI) and Land Surface Temperature (LST) (August 2022 to May 2023), and finally collected architectural information from the dwellings investigated. Our results show that the climatological data from the two communities are statistically different, being more favorable to comfort in the community with greater vegetation conservation. Analysis of remote sensing images showed that vegetation density in this community indicates a strong relationship with vegetative regeneration capacity after the most critical dry months of the period investigated. Regarding thermal comfort inside dwellings, the data showed that, both through statistical analysis and the thermal comfort model adopted, the combined effect of building materials, vegetation cover, and environmental context favors better thermal comfort conditions. The results presented point to the importance of vegetation conservation, which positively influenced the local climate and thermal comfort of riverside dwellings. In addition, our results encourage the perpetuation of traditional knowledge that adapts architecture to the forest.

Implementation and acceptability of high efficiency particulate air filters to reduce respiratory infections in care homes: Process evaluation of the AFRI-c cluster randomised controlled trial

Respiratory infections are easily transmitted within care homes. Within a clinical trial (called AFRI-c), which tested the effectiveness of high-efficiency-particulate-air (HEPA) filters to reduce respiratory infections in care home residents, we conducted a mixed-methods process evaluation. We aimed to understand their acceptability, fidelity, and implementation to aid interpretation of the effectiveness findings. We used qualitative remote and face-to-face interviews with staff (n = 25), residents (n = 20), and relatives (n = 12) from care homes (n = 22) in the AFRI-c trial. We purposively sampled homes for variation in size, nursing or residential provision, and deprivation. We used reflexive thematic analysis, drawing on normalisation process theory to understand implementation. We used staff questionnaires (n ranges from 191 to 351 depending on questionnaire) and resident (n = 1158) questionnaires with descriptive and regression analyses. We used the triangulation protocol to integrate qualitative and quantitative findings. The use of HEPA filters became normalised, although some residents disliked the draught. Self-reported intervention fidelity was high, which is important context for interpreting the trial’s null outcome for infection reduction. HEPA filters made no difference to resident or staff satisfaction with the care home environment. We found no evidence that using HEPA filters changed infection control and prevention practices. While staff felt it was a priority to prevent respiratory infections, residents were more concerned about quality of life and care. Our mixed methods process evaluation of the AFRI-c trial found the use of HEPA filters was acceptable, with high levels of adherence and low levels of contamination, suggesting that the null trial results were not due to poor adherence. Some effort was required to ensure they were kept on. Approaches to data collection may have caused under-reporting of mild infections. We should not assume that infection prevention is always a priority for residents.

The association between best friends or fixed peer groups and academic performance among Korean children and adolescents: a cross-sectional study

Selective GPR17 antagonism enhances structural and functional recovery in animal models of demyelination

Myelination, driven by differentiation of oligodendrocyte precursor cells, is critical for metabolic and structural support and efficient axonal signal transmission in neurons. Loss of myelin is a hallmark of multiple sclerosis and other devastating demyelinating disorders. As demyelination persists, neurons become increasingly vulnerable, leading to neurodegeneration and chronic disability. Restoring myelin through endogenous repair mechanisms offers a promising therapeutic approach to mitigate progressive neuronal loss. One key regulator of myelination is the G protein-coupled receptor 17, GPR17, whose chronic upregulation in oligodendrocyte precursor cells is commonly seen with myelin injury. In line with single-nucleus transcriptomic data showing predominant expression of GPR17 in committed oligodendrocyte precursor cells, our postmortem immunohistochemical analyses of MS patient tissue revealed a significant upregulation of GPR17 + /BCAS1 + oligodendrocyte precursor cells adjacent to and in demyelinated lesions. Importantly, remyelinated lesions lacked GPR17 immunoreactivity, consistent with a model in which sustained GPR17 expression is associated with demyelination and impaired oligodendrocyte precursor cell differentiation. To test the impact of pharmacological GPR17 inhibition on remyelination, we evaluated the effects of a novel, selective GPR17 antagonist in cuprizone-induced murine demyelination models. This toxin-induced approach has been widely used to study mechanisms of de- and remyelination, in the absence of the full inflammatory complexity of demyelinating diseases such as multiple sclerosis. We show that oral treatment results in robust functional recovery consistent with remyelination, as evidenced by improved spatial memory and recovery of visual evoked potential latency delays. GPR17 antagonism also accelerated structural remyelination in the corpus callosum and optic nerve. Together, these findings support a role for pharmacological GPR17 antagonism in promoting remyelination and highlight this G protein-coupled receptor as a promising therapeutic target for demyelinating disorders.

Accelerometer measured physical activity and menstrual distress across menstrual and nonmenstrual days in mild and moderate to severe primary dysmenorrhea

Abstract Dysmenorrhea and menstrual features have a major impact on patterns of physical activity; however, objective assessments of these associations using accelerometry remain limited. This study aimed to compare sitting and physical activity times, measured using a Fibion accelerometer, during the first three days of menstruation and days 8–10 of the follicular phase in age-matched women with mild and moderate-to-severe primary dysmenorrhea (PD). This observational study included 46 women aged 18–45 years whose primary dysmenorrhea severity was categorized based on the Numeric Pain Rating Scale (NPRS). Symptom severity was measured using Working ability, Location, Intensity, Days of pain, Dysmenorrhea (WaLIDD) and Dysmenorrhea Symptom Interference (DSI) scores. Sitting time and light, moderate, and vigorous physical activity were measured using a Fibion triaxial accelerometer worn for three consecutive days in the menstrual phase (days 1–3) and three consecutive days in the follicular phase (days 8–10). Multiple 2 × 2 mixed ANOVA analyses were used to assess differences between the menstruation and follicular phases in both groups for the outcomes of interest. Chi-square tests were performed to compare self-reported menstruation information. Participants demonstrated similar activity patterns during their menstrual and follicular days and spent most of their time sitting (~ 11 h/day), followed by light-intensity activity (~ 3.8–4.0 h/day), with considerably less time in moderate activity (~ 34–36 min/day) and minimal vigorous activity (< 1 min/day). No significant differences in sitting time (p = 0.586), light intensity (p = 0.414), moderate intensity (p = 0.776), or vigorous intensity (p = 0.162) were detected between the two dysmenorrhea groups across both phases. WaLIDD scores were not significantly different for the severity of pain over the last 3 months or the duration of menstrual pain but showed a statistically higher number of pain locations and level of interruption of daily activities in women with moderate-to-severe PD. The DSI differed significantly between the groups in all domains (p < 0.001). While objectively measured physical activity levels did not significantly differ between the menstrual and follicular phases or across dysmenorrhea severity groups, women with moderate-to-severe dysmenorrhea experienced greater symptom interference and disruption of daily activities. According to these results, dysmenorrhea significantly affects self-reported function and the perceived ability to be active during menstruation, even though it may not significantly change overall activity intensity. To clarify the physiological and behavioural mechanisms underlying these relationships, further longitudinal studies are needed.

Dual-input deep learning system for microbial identification from blood agar plates

Objective The morphological classification of microbial cultures and colonies requires specialized knowledge and experience; therefore, automation in this field remains limited. This study examines the feasibility of automating the identification of pathogenic microbial species from colony images of cultured microorganisms. Methods Two distinct image datasets were constructed for 10 clinically relevant species: a colony image dataset, consisting of individual colony images cropped, and a tile image dataset, generated by dividing entire images of the culture plate into tiles. Separate ResNet-50–based models were trained in each dataset, and their outputs were integrated to evaluate the classification performance. Training was conducted using 10,048 colony images and 23,003 tile images collected from 418 strains, and performance was assessed with five-fold cross-validation (K = 5). Results The colony image model achieved a sensitivity of 0.934 and a specificity of 0.993, while the tile image model achieved a sensitivity of 0.918 and a specificity of 0.991. Integration of the two models into an ensemble model further improved performance, yielding a sensitivity of 0.955 and a specificity of 0.995 when tested on 76 independent strains. Conclusion The ensemble model approach provides high accuracy and robustness, suggesting its potential technical contributions to microbial identification in clinical microbiology laboratories.

Integrative multi-omics characterization of HECTD3 across pan-cancer and its functional validation in thyroid cancer

Expression of Concern: Signatures of tumor microenvironment-related genes and long noncoding RNAs predict poor prognosis in osteosarcoma

Effect of massed versus distributed preclinical training on subgingival instrumentation skill across the preclinical-to-clinical transition

Abstract Consolidation of preclinical subgingival instrumentation skills during early clinical exposure is poorly quantified, especially after the new German dental licensing regulation (nGDLR). Within a shared digitised training programme (DTP), we compared in vitro performance after preclinical training and after the first periodontal patient course under distributed (oGDLR; 7th semester, S7) versus massed (nGDLR; 6th semester, S6) schedules. In a non-randomised, sequential two-cohort design (a consequence of a 2020 curriculum change), 102 students (oGDLR, distributed/longitudinal schedule, n = 45; nGDLR, massed/block schedule, n = 57) provided 136 evaluations after preclinical DTP (nGDLR S6, oGDLR S7) and/or after the first periodontal patient course (S8). Because oGDLR training (S7) overlapped with first clinical contacts whereas nGDLR training (S6) preceded intensive patient care by ~ 1.5 semesters, schedule density and clinical proximity co-vary and cannot be fully separated. Only 34 students (oGDLR n = 22; nGDLR n = 12) were evaluated at both time points, so most comparisons are cross-sectional. Each student instrumented six teeth with Gracey curettes (GRA) and a sonic scaler (AIR) on periodontitis manikins. Effectiveness of simulated plaque removal (ESP), treatment time, calculus removal and an 8-item Likert self-assessment were analysed by non-parametric tests and multivariable regression with cluster-robust SE. oGDLR was associated with higher ESP (β=+5.16%, p = 0.001) and twofold higher calculus removal (adjusted OR 2.00, p < 0.001); treatment time was comparable. From post-DTP to S8 both cohorts became faster by ~ 60s/tooth (p < 0.001) while ESP declined (β=−4.26%, p = 0.002) — a speed–accuracy trade-off; the schedule×time interaction was non-significant. In the paired subgroup, the trade-off was confirmed in oGDLR (n = 22; ESP − 7.2%, p < 0.001) but power-limited in nGDLR (n = 12). Excluding magnifying-loupe users (concentrated in nGDLR) widened rather than narrowed the oGDLR ESP advantage (β=+6.02%, p < 0.001). nGDLR rated AIR more favourably (learnability OR 0.40; effectiveness OR 0.48). Pass rates (≥ 55% ESP) exceeded 86%. In this non-randomised cohort, the distributed schedule was associated with higher absolute performance, but the decline occurred similarly under both schedules. Because schedule density could not be separated from proximity to clinical exposure, and because the data are largely cross-sectional and derive from a manikin-based in vitro model, these associations should not be read as causal. The transition to first patient care is vulnerable: cleaning quality may erode as students prioritise speed. Distributed preclinical practice was associated with a higher absolute baseline, but the decline across the transition was similar in both schedules; structured reinforcement during the early clinical phase remains necessary to consolidate competence.

Research on Benggang identification and deformation monitoring based on optical and Radar data

Benggang erosion, a severe soil erosion landform in southern China, threatens ecological security and regional development. Conventional methods can delineate surface morphology but lack capability to characterize vertical dynamics. To address it, this study presents an integrated method of deep learning and multi-temporal InSAR based on optical and radar data, and conducts benggang identification and surface deformation monitoring in typical benggang regions of Wuhua County, Guangdong, China. The results show that a U-Net model trained on Gaofen-2 high-resolution imagery achieved accurate automated Benggang delineation with a mean IoU of 86%. Subsequently, 90 Sentinel-1 SAR scenes acquired between 2022 and 2024 were processed using SBAS-InSAR and D-InSAR techniques, with Kriging interpolation employed to generate a spatially continuous deformation field. The framework successfully resolved millimeter-scale interannual vertical surface deformation, with internal cross-validation metrics (R 2  = 0.978, RMSE = 0.544 mm, MAE = 0.313 mm) confirming high spatial consistency of the fused deformation field. The proposed framework offers a reliable and scalable technical pathway for stereoscopic Benggang monitoring and demonstrates strong potential for geohazard early-warning and ecological risk management.

Reflux ratio modulates algal and bacterial metabolic balance to enhance refractory pollutant removal

Incidence of tuberculosis in patients with immune-mediated diseases undergone biological therapy: A 10-year observational study in a high-burden region of northeastern Brazil

Biological therapies are critical in the management of Immune-Mediated Diseases (IMDs), yet their use is linked to adverse effects, including an elevated risk of Tuberculosis (TB), a significant health issue in endemic regions like Brazil. This study aimed to assess the incidence of TB among patients with IMDs in rheumatology, gastroenterology, and dermatology who received biological therapy in Pernambuco between 2013 and 2023. A total of 8,623 patients were analyzed, with 90.7% diagnosed with rheumatologic diseases, 5.6% with gastroenterological diseases, and 3.7% with psoriasis. In the descriptive analysis, frequency distributions and medians with interquartile ranges were reported, and Relative Risk (RR) was calculated to compare tuberculosis occurrence across IMD groups. The overall TB incidence was 2.1%, highest in gastroenterological patients (4.7%) compared to rheumatologic (2.0%) and psoriasis patients (1.3%) with a risk ratio (RR) of 2.4 for the group of gastroenterological diseases (p < 0.001). No significant differences in TB frequency were detected before (1.16%) or during (0.97%) the use of biological therapy, nor among drug classes, except for a slightly higher incidence with infliximab (2.6%). The absence of differences between the other anti-Tumor Necrosis Factor (anti-TNF) drugs and other cytokine inhibitors suggests that TB incidence might be influenced by environmental exposure in endemic regions. Pulmonary TB was the predominant form (80.8%), and pleural TB emerged as the most frequent extrapulmonary manifestation. The results underscore the need for extensive regional studies to elucidate TB risks in IMD patients and refine treatment guidelines for immunobiological therapies.

Infrared thermal imaging to determine temperature parameters of dengue vector habitats across ecological regions of Nepal: a pilot feasibility study

Abstract Dengue is a major vector-borne disease in Nepal, driven by container-breeding Aedes mosquitoes that thrive in domestic, urban settings. Its risk is increasing from the lowland Terai to the Hill and Mountain regions due to urbanisation and global warming. Practical and innovative approaches are needed to better understand the impact on Aedes mosquito habitats that are the main targets for vector control. We conducted a pilot feasibility study to characterise Aedes aegypti and Aedes albopictus breeding sites, examine physicochemical parameters (total dissolved solids, pH, salinity, dissolved oxygen), and explore the utility of infrared thermal imaging as a novel tool for measuring temperature parameters of habitats across distinct ecological regions. We found that plastic containers, tin/metal items, and rubber tyres were common breeding sites. The dengue vectors Ae. aegypti and Ae. albopictus were identified in a variety of container types across most locations. Overall, there was reasonable agreement between the thermal camera and standard thermometer, although the camera recorded approximately 1 °C lower and showed divergence at higher temperatures. Most aquatic habitat temperatures were within an optimal mosquito breeding range (mid-20s to low 30s °C) and lower than non-breeding site temperatures. The breeding site water temperatures in the Terai region were ~ 4 °C higher than the Mid-Hill and Mountain regions. Aquatic habitat temperatures were also found to be significantly correlated with dissolved oxygen. Thermal imaging case studies provided key visual insights into varying temperatures in complex Aedes microclimates. Notably, inner rubber tyre temperatures remained relatively stable despite significant external temperature fluctuations and external tyre temperatures exceeding 50 °C. Thermal imaging shows promise for the field characterisation of mosquito breeding sites, but it is not interchangeable with standard thermometry without qualification, including calibration and uncertainty analysis.

Evaluating the cultural alignment of multilingual LLMs in typical Japanese workplace scenarios

While current evaluations of LLM cultural alignment predominantly rely on static benchmarks in Western contexts, their ability to navigate generative, high-context socio-pragmatic demands in non-Western environments remains critically underexplored. This study investigates how multilingual LLMs adapt to the Japanese workplace—a stringent stress-test environment characterized by strong high-context communication norms and rigid honorific conventions—using Hofstede’s six cultural dimensions as a heuristic framework. We evaluated five state-of-the-art LLMs (LLM-jp, Phi, Llama, Qwen, and GLM) through a large-scale crowdsourced human evaluation. Based on 1,718 valid evaluator sessions, native Japanese raters assessed model outputs to generate a holistic Japanese Workplace Cultural Alignment Score (JWCAS). To dissect the underlying communicative strategies, we paired this with a three-layer diagnostic sub-score analysis (Linguistic Form, Socio-Cultural Values, and Social Action). Our results reveal that leading multilingual models (Phi and GLM) achieved overall JWCAS scores comparable to, or significantly higher than, the native Japanese model (LLM-jp). Crucially, our sub-score analysis demonstrates that holistic evaluation metrics can obscure deep pragmatic deficits: while LLM-jp overfits to surface-level linguistic politeness (Layer 1), it shows critical weaknesses in socio-cultural values (Layer 2) and context-aware social strategies (Layer 3). In contrast, leading multilingual models demonstrate balanced competence across all layers. These findings suggest that true cultural competence requires moving beyond native linguistic mastery, highlighting the necessity of multi-dimensional diagnostic frameworks for cross-cultural AI alignment.

Milk yield and immune protein profiles in colostrum and milk of indigenous syrian Golani and Golani × Friesian crossbred cows

PAP_NER: A large-scale vietnamese administrative named entity recognition corpus and hybrid deep learning architecture

Named Entity Recognition (NER) is fundamental for automating administrative document processing in digital government systems. However, Vietnamese NLP research faces a critical infrastructure gap: existing datasets focus on generic information extraction (news, medical) rather than domain-specific administrative text. We present PAP_NER, the first large-scale, gold-standard Vietnamese administrative NER corpus comprising 162,801 sentences with 205,807 entity annotations across five entity types critical for e-Government workflows: Agency (CQ), Legal Document (VBPL), Object (ĐT), Datetime (NG), and Quantity (SL). The dataset was constructed through a rigorous human-in-the-loop annotation pipeline, achieving an inter-annotator agreement of κ = 0.85. We demonstrate PAP_NER’s value through comprehensive benchmarking of an established hybrid deep learning architecture, PhoBERT-CRF, which couples monolingual Transformer embeddings (PhoBERT) with Conditional Random Fields for structured prediction. PhoBERT-CRF achieves 97.95% Micro F1-score on the PAP_NER test set, significantly outperforming established baselines: BiLSTM+CRF (+2.01%), multilingual XLM-RoBERTa (+2.52%), and pure Transformer approaches (+0.44%). Ablation analysis reveals that the CRF layer provides statistically significant improvements for structurally complex entities (VBPL: + 0.96%, p  < 0.05, McNemar’s test). We release PAP_NER publicly (DOI: 10.5281/zenodo.18044019 ) under Creative Commons BY 4.0 license to support reproducibility and enable further research in Vietnamese administrative NLP. This work establishes a foundational dataset and methodology for addressing the Vietnamese government NER gap, with implications for low-resource language NLP research.

Importance of overlapping network nodes in influence spreading