Browse Articles

Study on elastic-plastic displacement response spectra of base-isolated structures and its influencing factors

In seismic design, the acceleration response spectrum method is used to calculate the bearing capacity of structures to ensure their safety. However, under intense earthquake action, insufficient deformation capacity of structures is the main cause of their failure. For base-isolated structures, the failure is caused by the displacement of the isolation layer exceeding the limit value. To derive practical elastic-plastic displacement spectra for base-isolated structures and incorporate them into seismic design, this paper categorizes 1217 strong motion records, with peak accelerations exceeding 10 gal, into nine groups based on soil conditions and earthquake type. To make the obtained displacement response spectra more accurate, this paper establishes the motion equation of a two-degree-of-freedom system model, using non-proportional damping and the Bouc-Wen model to describe the damping and mechanical characteristics of the isolation layer, the elastic-plastic displacement response spectrum is calculated using numerical methods. Accordingly, a practical formula for the elastic-plastic displacement spectra of base-isolated structures is established by analyzing the factors influencing (soil conditions, earthquake type, parameters in Bouc-wen model) the spectra. Herein, the elasto-plastic displacement response spectrum for base-isolated structures was established, and a formula for calculating the displacement of the isolation layer in base-isolated structures with natural vibration periods within 10 seconds was provided, along with the design peak ground displacement under different conditions. Based on the elastic-plastic analysis of two distinct base-isolated structures, results demonstrate that the current formula is an effective and rapid predictor for estimating the displacement of the isolation layer in base-isolated structures considering various site conditions.

Correction: A two-stage forecasting model using random forest subset-based feature selection and BiGRU with attention mechanism: Application to stock indices

Optimal colors can predict luminosity thresholds in natural scenes

Luminosity thresholds define the luminance boundary at which a surface color shifts in appearance from being perceived as an illuminated surface to appearing self-luminous. Previous research suggests that the human visual system infers these thresholds based on internal references of physically realizable surface colors under a given illumination, referred to as the physical gamut. A surface is perceived as self-luminous when its luminance exceeds the upper limit of this empirically internalized gamut. However, the precise structure and boundaries of these gamuts remain uncertain. Optimal colors, which represent theoretical surface reflectances under specific illuminants, have been shown to provide an effective model for visualizing and computing the physical gamuts. In prior studies, optimal colors have successfully predicted luminosity thresholds; however, these findings were limited to highly simplified, abstract stimuli. Whether this framework generalizes to more naturalistic viewing conditions has remained an open question. In the present study, we demonstrate that the theory of an internal reference in the form of an empirically constructed physical gamut, visualized through optimal colors, remains valid under more natural conditions. Our results confirm that optimal colors can still accurately predict luminosity thresholds in such settings. Moreover, our findings suggest that the luminosity thresholds encompass both self-luminosity and naturalness concepts. Consequently, this may imply that the notion of physical gamut could encompass both concepts as well and could be defined as “all physically possible colors in a scene for an object that does not emit light.” These insights have profound potential implications for applied fields (e.g., XR or projection mapping). For example, they provide a theoretical framework that specifies the luminance and color boundaries virtual objects should not exceed to remain realistically integrated into AR/MR scenes. In fundamental science, these findings can improve our understanding of the perception of naturalness in images.

Effects of protein sources at sahur on anaerobic power and strength during Ramadan in combat sport athletes: A single blind, randomized, placebo-controlled, counterbalanced crossover study design

This study investigated the acute effects of different protein sources consumed at sahur on anaerobic power and strength performances in trained male combat sport athletes during Ramadan fasting. Using a single blind, randomized, placebo-controlled, counterbalanced crossover study design, 24 male combat sports’ athletes (mean age: 27.3 ± 3.8 years, Tier 3 national level) completed four experimental conditions: (1) non-fasting control, (2) fasting + placebo (maltodextrin), (3) fasting + whey protein isolate (WPI), and (4) fasting + micellar casein (MC). In each condition, a standardized sahur meal (6.3–7.7 kcal/kg body weight) and supplementation (0.4 g/kg for WPI/MC and 0.4 g/kg for Placebo) were administered. Physical Performances was assessed 11–13 hours post-sahur (or 3–5 hours post-lunch for control) including the Wingate anaerobic test, bench press, leg press, and countermovement jump (CMJ), and handgrip strength tests. Ramadan fasting significantly lowered Wingate peak power, mean power, and bench press strength compared to the non-fasting control. MC supplementation reduced these declines, outperforming WPI and the placebo in peak power and mean power, and surpassing the placebo in bench press strength, although not WPI. Leg press, countermovement jump, and handgrip strength showed no significant differences across conditions. MC supplementation at sahur provides partial protection against fasting-induced declines in anaerobic power and upper body endurance, but does not fully restore performance to non-fasting levels. These findings emphasize the importance of protein timing and selection in mitigating performance decrements during Ramadan fasting, highlighting the need for further research on optimal nutritional strategies for athletes training and competing under fasting conditions.

An interpretable vibration-enhanced BRB model for rolling bearing fault diagnosis

The operational condition of rolling bearings is essential to the reliability of industrial machinery, making fault diagnosis a critical research topic. Although deep learning has gained widespread attention in this domain, its black-box character and reliance on a large number of training samples limit its practical applicability. In contrast, knowledge-driven intelligent diagnostic models have gained increasing interest due to their superior interpretability and robustness under small-sample conditions. The Belief Rule Base (BRB) model is a representative example of such interpretable methods. However, the conventional BRB models struggle to process continuous signals, limiting their effectiveness in real-world bearing fault diagnosis. This study proposes a novel Vibration-Enhanced Belief Rule Base (VE-BRB) model designed to address this limitation. First, the window feature extraction method is used to preprocess the continuous vibration signal. Bearing fault features are extracted from low-frequency and high-frequency to construct an energy matrix representation. Thereby, the original continuous vibration signal can be effectively mapped to the rule matching space. Second, the model is reasoned through an evidential reasoning (ER) algorithm. This guarantees the interpretability of the model. Finally, the projection covariance matrix adaptation evolution strategy (P-CMA-ES) is employed as the optimization process. To validate the effectiveness of the proposed method, the test was performed using bearing datasets from Case Western Reserve University and Huazhong University of Science and Technology under various conditions.

Biological evaluation of amidine derivatives: In vitro cytotoxicity and cellular antioxidant capacity

Amidines and related compounds are well known intermediates and protecting groups in organic synthesis. New methodological approaches and obvious structural and functional relevance to guanidines and imidazoles have also prompted interest in the biological activity of these compounds. Here we report a preliminary cytotoxicty evaluation of a set a formamidines and formamidine ureas obtained by convenient and modular synthetic routes. Standard epithelial (Vero, MDCK-SIAT) and fibroblast cell lines (COS-1, COS-7) were employed. All compounds were found to be relatively non-toxic, with LC50 values all in excess of 0.3 mM, but found to vary over the range of compound structures. Cell morphological changes were in good agreement with cell viability. Most of the compounds either suppressed the cellular antioxidant capacity or promoted reactive oxygen species (ROS) generation. The nontoxic nature of these molecules at low to moderate concentrations suggests that the amidine and formamidine urea functional groups are suitable for continued investigation in drug development.

Phytochemical profiling of Vitex negundo seeds via UHPLC-QTOF-MS/MS analyses with antimicrobial evaluation and in silico targeting of DNA Gyrase B and Secreted Aspartic Proteinase 2 (SAP2)

The present study reports the metabolic profiling and antimicrobial evaluation of Vitex negundo seed extract. UHPLC-QTOF-MS/MS analysis identified seventeen bioactive phytoconstituents, correlated with the observed antimicrobial and antifungal activities. Among them, isoorientin (–7.3 kcal mol ⁻ ¹), quercetin (–7.8 kcal mol ⁻ ¹), and orientin (–7.4 kcal mol ⁻ ¹) exhibited strong binding affinities towards Staph Gyrase B (24 kDa). Similarly, isoorientin (–8.1 kcal mol ⁻ ¹), quercetin (–8.4 kcal mol ⁻ ¹), and orientin (–8.3 kcal mol ⁻ ¹) displayed significant interactions with secreted aspartic proteinase (SAP2) enzyme, confirming their antimicrobial potential. The aqueous-methanolic seed extract demonstrated notable inhibitory activity against Staphylococcus aureus (26.4 ± 0.3 mm; 44.08% inhibition) and Candida albicans (25.7 ± 0.4 mm; 29.73% inhibition). Density functional theory (DFT) calculations at B3LYP/6-31G level were used to optimize the ground state geometries of the identified phytochemicals and analyze their frontier molecular orbitals (FMOs) and global reactivity descriptors. Time-dependent DFT (TDDFT) calculations at the B3LYP/6-311G level (solvent: DMSO) further explored their biological relevance and nonlinear optical (NLO) properties, including ionization potential (IP), molecular electrostatic potential (MEP), and HOMO-LUMO energy gaps. These quantum chemical parameters provided mechanistic insights into the antimicrobial potential of the identified constituents. Molecular docking simulations further confirmed strong geometric complementarity and favorable binding affinities, highlighting the Vitex negundo seed extract as a promising source of a novel medicinal agent with previously unreported antifungal and antibacterial activities.

Correction: Youth focused life skills training and counselling services program–An inter-sectoral initiative in India: Program development and preliminary analysis of factors affecting life skills

Development of InDel markers associated with leaf protein content in mulberry based on whole-genome resequencing

Mulberry ( Morus alba L.), as a crucial dual-purpose crop for medicinal and forage applications, requires genetic improvement of leaf protein content to substantially enhance its feed value. This study utilized 46 mulberry germplasms as test materials, with 29 accessions subjected to whole-genome resequencing on the Illumina HiSeq platform. Based on the resequencing data, Indel primers were designed and screened to analyze genetic diversity and develop Indel markers tightly linked to crude protein content in mulberry leaves. The results showed that a total of 1 155 585 InDel loci were detected in the 29 accessions, with 11 401 InDels located in coding regions (0.99%). Analysis of functional annotations in extreme-phenotype germplasm revealed that 2 563 InDel-containing genes within coding sequence (CDS) regions of high-protein accessions were significantly enriched in catalytic activity and metabolic processes. Further utilizing InDel variations within CDS regions, 98 pairs of InDel primers were designed, yielding a novel linked marker (Chr1-3-204) with distinct amplification patterns between high- and low-protein accessions. Validation in 46 accessions showed 89.1% accuracy (r = 0.726, P < 0.01). Sanger sequencing confirmed its high accuracy (97.7% similarity to reference), with 84% identity to Morus motabilis . This study is the first to develop a linked marker for crude protein content in mulberry leaves, providing a technical reference for the breeding of high-protein mulberry varieties.N

Beyond the grid: Navigating water supply and sanitation service ecosystems in informal settlements

Hundreds of millions of people living in urban informal settlements rely on irregular and unsafe water supply and sanitation services. To meet their needs, they must navigate fragmented service delivery environments and use multiple different water and sanitation facilities. Using high-frequency longitudinal survey data from three informal settlements in Kenya, Peru and South Africa, we document the variability in water supply and sanitation service access. Across the year-long study period, 62–73% of respondents across all contexts changed their primary toilet, with 10–27% reporting five or more different primary facilities, with similar variability in water access. High levels of disruption were reported, with issues related to crowding/queuing, breakdowns, and physical barriers disrupting accessibility all contributing to the churn of services. To explain the results, we develop the concept of a “service ecosystem” to describe how people living in urban informal settlements rely on multiple water supply and sanitation services simultaneously and how these patterns of access shift over time. Using James C Scott’s theory of legibility, we argue that this irregularity means these service ecosystems are largely illegible within formal monitoring frameworks, that typically categorise households by a primary service. This leads to an information deficit for policymakers and practitioners who have a mandate to improve services in these environments. We further develop the implications of service ecosystems by calling for policymakers and service providers to recognise and support a diversity of service systems which have sufficient redundancy between them to meet the needs of populations, at least until broader structural reforms can address the underlying challenges in these settings.

Machine learning-driven optimization of monolithic gold plasmonic sensors: Achieving ultrahigh sensitivity with interpretable linear models

Integrating machine learning (ML) with nanophotonic engineering, this work achieves unprecedented performance in surface plasmon resonance (SPR) biosensing through a co-designed gold-coated photonic crystal fiber (PCF-SPR) sensor and multi-algorithm computational framework. An asymmetric circular PCF structure with concentric air-hole rings ( Λ 1 = 3.26   μ m , Λ 2 = 2.12   μ m ) and a 50 nm gold layer maximizes evanescent field-analyte overlap, generating complex spectral signatures ideal for machine learning interpretation. High-fidelity COMSOL Multiphysics simulations produce 1560 synthetic data points across refractive indices (RIs) of 1.33–1.38, capturing confinement loss, wavelength sensitivity, and effective permittivity. Three regression models—Multiple Linear Regression (MLR), Support Vector Regression (SVR), and Random Forest Regression (RFR)—are rigorously evaluated for predicting optical responses. The sensor demonstrates a record wavelength sensitivity of 31 846.46 nm/RIU -1 at R I = 1.33 , with minimal variation (0.02%) across the biological range, alongside a resolution of 1.57 × 10 − 3 RIU. Crucially, MLR outperforms nonlinear counterparts, achieving superior accuracy in confinement loss (MAE = 3.97, RMSE = 5.03) and sensitivity prediction (MAE = 40.18, RMSE = 50.54). This synergy of optimized pure-gold microstructures and interpretable machine learning establishes a robust pipeline for high-sensitivity, noise-resilient biosensing, surpassing prior ML-enhanced plasmonic sensors in critical performance metrics while simplifying fabrication.

Forecasting emergency department visits in the reference hospital of the Balearic Islands: The role of tourist and weather data

Accurate forecasting of patient arrivals at emergency departments (EDs) is vital for efficient resource allocation and high-quality patient care. In this study we investigate the relevance of exogenous variables, namely tourism, weather, calendar and demographic variables, in forecasting ED visits in the reference hospital in Palma de Mallorca, a city with significant seasonal population fluctuations due to tourism. Using a machine learning approach, we develop a model that predicts ED visits based solely on these exogenous variables. We test different machine learning algorithms (random forests, support vector machines, and feedforward neural networks) with different combinations of input variables and compare their symmetric mean average percentage errors (SMAPEs). Our findings reveal that calendar information, resident, and tourist population data are statistically significant for the accuracy of the predictions, while the addition of weather data does not provide any further improvement. Comparison of non-time-series with time-series prediction models reveals that the latter provide better accuracy for short prediction horizons (e.g., shorter than a week). Furthermore, time-series models become less or equally accurate to models relying only on exogenous variables for long prediction horizons (e.g., fortnight or month). Our study highlights the importance of carefully selecting predictive variables to ensure short- and long-term, robust and reliable forecasts. This demonstrates that, despite their lower complexity, non-time-series models with well-chosen input variables can be as effective as time-series models when predicting for long time horizons.

Effect of iodine nutrition on blood glucose and blood lipid levels in different regions: Based on quantile regression

Objective Iodine is essential for the synthesis of thyroid hormones in the human body, and both excessive and insufficient iodine intake can significantly impact metabolic processes and contribute to various diseases. This study aims to investigate the effects of serum iodine levels and other associated factors on blood glucose and blood lipid levels. Methods A total of 1344 participants were recruited from three distinct regions in Shandong province, each characterized by varying levels of environmental iodine content. Blood and urine samples were collected from each participant and analyzed for iodine nutrition status, blood glucose (BG), blood lipids, and other relevant biochemical indicators. Quantile regression analysis was employed to examine the continuous effects of serum iodine concentration (SIC) and other related factors on BG, total cholesterol (TC), triglycerides (TG), high-density lipoprotein (HDL), and low-density lipoprotein (LDL). Results In men ≥ 45 years, higher SIC was significantly associated with reduced TG at the 75th (β = −0.015, 95%CI: −0.030, 0.000) and 90th (β = −0.035, 95%CI: −0.064, −0.006) percentiles ( P  < 0.05). In men < 45 years, SIC affected TC at the 25th percentile (β = 0.008, 95%CI: 0.001, 0.015; P  < 0.05). In women ≥45 years, SIC increased BG at the 90th percentile (β = 0.006, 95%CI: 0.001, 0.011), TC at the 50th (β = 0.005, 95%CI: 0.002, 0.009) and 75th (β = 0.005, 95%CI: 0.001, 0.009) percentiles, TG at the 50th percentile (β = 0.004, 95%CI: 0.001, 0.007), and HDL-C at the 25th percentile (β = 0.001, 95%CI: 0.000, 0.001) (all P  < 0.05), but had no significant effect on LDL-C. In women <45 years, SIC only increased HDL-C at the 75th (β = 0.005, 95%CI: 0.001, 0.009) and 90th (β = 0.005, 95%CI: 0.001, 0.009) percentiles ( P  < 0.05). Conclusion SIC exerts gender- and age-specific effects on BG and lipids, particularly at upper percentiles of metabolic indicators in adults ≥45 years. These findings highlight the need for targeted iodine nutrition interventions to mitigate metabolic risks.

Longitudinal evidence of technology-enhanced, individualized neuromotor rehabilitation on autonomy, cognition, quality of life, and psychological well-being: Pilot multi-sample study

Background Over the last decades, neuromotor rehabilitation programs have integrated multidisciplinary approaches with the implementation of emerging technology (e.g., robotics and virtual reality – VR), to effectively target recovery complexity. While this strategy supported patient’s physical improvement, little evidence has been reported regarding the widespread effects on non-motor rehabilitation outcomes. Methods A prospective, two-arm, non-randomized study design was adopted to provide pilot feasibility evidence on the multi-domain impact of personalized technology-enhanced neuromotor rehabilitation from convenience sub-samples of patients with stroke, Parkinson’s Disease (PD), and osteoarthritis (OA). Technological intervention consisted of the integrated use of robot-assisted and/or VR-based exercises, individualized based on patient’s diagnosis and rehabilitation goals. Study outcomes included patient’s functional status (autonomy in ADLs, risk of falls), cognition (attention and executive functions, memory, verbal fluency), physical and mental health-related quality of life (HRQoL), and psychological status (anxiety and depression symptoms, and well-being) and were compared to patients participating in standard training only. Rehabilitation experience and technology psychosocial impact were also evaluated. Intra- and intergroup comparisons along with general linear models were statistically tested within each sub-sample considered independently over three timepoints (baseline, post-intervention, 6-month follow-up). Results At post-intervention, significant multi-domain intra-group improvements were observed within each sub-sample. Between-group differences were found on ADLs autonomy (stroke and PD; p  < .05), executive functions (stroke; p  < .01), anxiety and depression (OA and PD, respectively; p  < .05), and well-being (stroke and OA; p  < .05). Interaction effects (time x group) were significant only on well-being variables in stroke ( p  = .01) and OA ( p  = .02), evidencing wider short-term effects of technology-enhanced programs compared to standard training. At 6-month, significant time effects indicating sustained improvements over the three timepoints were estimated on HRQoL within each sub-sample ( p  < .05) and, additionally, on anxiety and depression in stroke ( p  = .02) and OA ( p  < .001). Interaction effects emerged only on physical HRQoL in OA ( p  = .02), along with significant between-group differences on HRQoL and anxiety and depression in OA ( p  < .05) and PD ( p  = .01), respectively. Conclusion Further full-scale trials are warranted to confirm the longitudinal trends observed in this pilot study and to further investigate the potential multi-domain benefits of multidisciplinary and technology-integrated recovery approaches across different clinical populations. Trial registration ClinicalTrials.gov ID: NCT05399043 .

Elucidating the mechanistic association of xylene inducing non-small cell lung cancer through network toxicology and molecular docking analysis

Xylene is a common industrial solvent that includes three isomers: o-xylene, m-xylene, and p-xylene. Long-term exposure to low doses of xylene in the environment has been linked to a higher risk of lung cancer. However, the molecular mechanisms behind this link are still not fully understood. In this study, we used a combination of network toxicology and molecular docking to investigate how xylene may contribute to the development of non-small cell lung cancer (NSCLC). We first identified 115 potential target genes related to xylene exposure by searching several public databases, including CHEMBL, STITCH, GeneCards, and OMIM. Further screening using the STRING platform and Cytoscape analysis highlighted five core targets: IL1A, H3C13, ITGAM, CCR5, and COMT. We utilized scRNA-seq data to analyze the expression patterns of core targets across distinct cell subpopulations, the majority of core targets were expressed in immune cells. We then performed GO and KEGG pathway enrichment analysis. These results showed that the five target genes are mainly involved in cancer-related pathways, such as ECM-receptor interaction, focal adhesion, chemical carcinogenesis, and the PI3K-Akt signaling pathway. Molecular docking results confirmed that xylene isomers have strong binding affinities with the proteins encoded by these genes. This suggests that xylene may disrupt important cellular signals and promote tumor growth. In conclusion, our study provides new insight into how xylene might cause NSCLC at the molecular level. It also shows the usefulness of network toxicology in evaluating health risks from environmental chemicals. These findings may help guide future efforts in prevention and treatment strategies.

Capacity and site readiness for hypertension control program implementation in Nigeria: A nationwide cross-sectional study

Background The aim of this study is to determine the capacity and readiness of Nigerian primary healthcare facilities to adopt a multi-level approach for the diagnosis, treatment, and control of hypertension. Methods Using a multi-stage sampling technique, 5 states were selected for the nationwide study, and 10 Primary Healthcare Centres (PHCs) were selected from each of the participating states. The 50 PHCs were evaluated using the World Health Organization-modified Service Availability and Readiness Assessment, focusing on the diagnosis and treatment of hypertension in Nigeria. The indicator scores for general and cardiovascular service preparedness were computed using the proportion of PHCs with accessible facilities, tools, diagnostic guidelines, and prescription drugs. Results A majority of PHCs (n = 43; 86%) reported having two or more full-time staff. The median number of full-time employees for the 50 PHCs was 6 (IQR = 2–9), and for the community health extension workers (CHEWs), the median was 2, the interquartile range (IQR) = 0–4. None of the PHCs had full-time physicians. Ninety-eight percent, 94%, and 84% of the 50 PHCs are able to provide screening services, diagnose, and confirm hypertension, respectively. In addition, 98% of the PHCs had functional blood pressure apparatus. However, only a minority of PHCs had the guidelines (24%), treatment algorithms (27%), and facilities. Most of the 50 PHCs (96%) use electronic patient records in their respective centres. Of the 50 PHCs studied, 66% had at least one 30-day antihypertensive treatment regimen in stock. The most commonly available drug classes were calcium channel blockers (72%), followed by diuretics (42%), central acting agents (38%), and angiotensin-converting enzyme inhibitors (36%). The median number of 30-day regimens in stock was 15 (IQR 0–132). Conclusion This first large-scale systematic assessment of capacity and readiness for a system-level hypertension control program within five states of Nigeria demonstrated implementation feasibility based on the workforce, equipment, and health information systems, but there is a critical need for health-worker training and provision of protocols for hypertension treatment and control, as well as some need to strengthen the essential medicine supply chain.

The dolutegravir failure cohort: A multi-country longitudinal cohort with a randomised clinical trial of continued dolutegravir versus switch to darunavir in people with viraemia while on dolutegravir in Sub-Saharan Africa (The Ndovu Study) protocol

Background There is insufficient data to inform the management of dolutegravir failure, with the WHO and various countries adopting different approaches, underscoring the need for an evidence-based management approach. Methods The Ndovu study is a large multi-country cohort, with a nested randomised controlled trial (RCT), enrolling 6,600 people living with HIV (PLWH) with viral load (VL) of ≥1000 copies/ml after at least 6 months of dolutegravir. Participants aged ≥ 1 year, including pregnant women, will be followed up for 12 months with enhanced adherence counselling (EAC) provided monthly. Viral load (VL) testing will be conducted every 3 months and drug resistance testing conducted if VL ≥ 200 copies/ml. Three hundred and sixty-two participants aged ≥15 years and 30 participants aged 3–14 years with major dolutegravir-associated drug resistant mutations (DRMs) will be enrolled into the RCT and randomised to switch to ritonavir boosted darunavir (DRV/r) or continue with dolutegravir with follow-up for 12 months. VL will be measured at 1, 3, 6 and 12 months and tenofovir levels assessed on dried blood spots at month 1 and month 6. The primary outcome of the cohort is the proportion of participants achieving viral load <200 copies/ml by month 12 and the primary endpoint of the RCT is viral load <200copies/ml at 6 months using a modified FDA snap shot algorithm. Secondary endpoints are DRM patterns associated with non-suppression, level of adherence associated with suppression as well as participant and provider experiences of staying on DTG versus switching to DRV/r. The RCT primary efficacy analysis will be conducted on the Intent-to-Treat Exposed (ITT-E) population and will compare the difference in the proportion of participants with viral load <200 copies/ml 6 months after randomisation between the treatment arms stratified by the randomisation stratification factors. This study is registered at ClinicalTrials.gov, NCT06762054 (cohort) and NCT06747507 (RCT) and enrollment into the cohort started in March 2025. Conclusion The Ndovu study will address critical gaps in the management of DTG failure including the emergence, determinants and implications of DTG resistance. Further, it will evaluate the optimal ART regimens to use in the setting of DTG resistance in adults and children.

The role of phase angle and standardized phase angle in assessing nutritional status and predicting complications in gastrointestinal cancer

Objective This study aims to look at the link between bioelectrical impedance analysis (BIA), especially phase angle (PA) and standardized phase angle (SPA), and how often postoperative infectious complications occur in patients with gastrointestinal tumors. Material and methods We chose 139 patients who had gastrointestinal tumor surgery. We did BIA tests on them, assessed their nutritional status with the Patient-Generated Subjective Global Assessment (PG-SGA) scale, checked body composition indicators, and tested their blood markers. The patients were categorized into high PA and low PA groups based on their PA values, and the differences in nutritional status-related indicators and infection rates between these groups were analyzed. Additionally, PA values were standardized, resulting in the formation of high SPA and low SPA groups, and further comparisons of nutritional indicators and infection rates between these groups were conducted. Results Our results show that PA and SPA are very important for checking the nutritional status of patients with digestive tract cancer. They are also closely linked to the chance of having postoperative complications. The study encompassed an analysis of 139 patients with digestive tract cancer, leading to several key conclusions: Firstly, a strong correlation was observed between PA and nutritional status, with malnourished patients demonstrating significantly lower PA values compared to those with adequate nutritional status. Secondly, the group with low PA was significantly associated with several adverse indicators, including advanced age, a higher proportion of females, increased prevalence of chronic diseases, lower BMI, elevated PG-SGA scores, higher incidence of sarcopenia, and reduced skeletal muscle mass and skeletal muscle index. Thirdly, patients categorized in the low SPA group exhibited a significantly higher incidence of complications relative to those in the high SPA group. Lastly, high SPA was significantly associated with a lower incidence of postoperative infectious complications, whereas TNM staging was significantly associated with a higher incidence of these complications. Conclusion This study substantiates the utility of PA and SPA in evaluating the nutritional status of patients with digestive tract cancer and in being associated with postoperative complications. PA demonstrates a significant correlation with malnutrition, sarcopenia, and various body composition indicators, while SPA is significantly associated with postoperative infectious complications.

Influence of environmental factors and tributaries on toxic cyanobacterial growth

Toxic cyanobacteria are widely present in water bodies worldwide, particularly in reservoirs. The Calima Reservoir is primarily used for energy generation, but some areas of it are also designated for recreational activities, making it a significant contributor to regional and national tourism. In this study, four sampling events were conducted in 2023, covering different seasonal periods. During these samplings, various species of potentially toxic cyanobacteria were detected; the presence of these cyanobacteria was analyzed in relation to several physicochemical variables, and microcystin‑LR (MC-LR) was detected in four of the six sampling stations, although in concentrations below the maximum permissible limit. The reservoir’s temperature showed a positive correlation with cyanobacterial abundance. One of the objectives of the research was to assess the growth of M. aeruginosa by exposing it to water samples from four tributaries that flow directly into the reservoir, each with distinct physicochemical properties. The highest growth induction, exceeding 400%, was observed in samples from Tributary 2 (T2), located at the center of the reservoir. The analysis revealed that the variables most significantly associated with enhanced M. aeruginosa proliferation were ammoniacal nitrogen (32.8 mg NH 3 -N/L), alkalinity (280 mg CaCO 3 ), and Kjeldahl nitrogen (33.6 mg N/L). In contrast, negative associations were observed with indicators of biodegradable organic matter, including of biochemical oxygen demand (BOD), chemical oxygen demand (COD), total organic carbon (TOC) and coliform-type microorganisms, indicating an environment conducive to the proliferation of M. aeruginosa . These findings suggest that it is important to conduct detailed analyses of tributary waters to identify areas with high potential for cyanobacterial blooms. These findings highlight the influence of external nutrient inputs on bloom formation and the need for focused mitigation to protect water quality.

Bayesian networks for predicting clinical outcomes in COVID-19 patients: A retrospective study in a resource-limited setting

Background The COVID-19 pandemic has highlighted the critical need for robust, interpretable predictive models to guide clinical decision-making for hospitalized patients, particularly in resource-limited settings. While machine learning approaches have achieved high predictive performance, many lack the transparency required for clinical adoption. Bayesian networks provide a rigorous mathematical framework for modeling medical uncertainty and complex causal relationships while maintaining clinical interpretability. Objective To develop and validate an interpretable Bayesian network model for predicting clinical outcomes in hospitalized COVID-19 patients, including severity, complications, and mortality, and to compare its performance with existing predictive approaches. The model specifically addresses the needs of resource-limited settings where both interpretability and performance are critical. Methods This retrospective cohort study analyzed 124 hospitalized COVID-19 patients at Tanambao I University Hospital, Madagascar (March 2020–March 2022). Given limited genomic sequencing capacity during the study period, specific SARS-CoV-2 variants infecting individual patients could not be confirmed. After descriptive and statistical implicative analysis, we constructed a Bayesian network integrating predictive variables organized in three levels: input variables (demographic, vital signs, biological, radiological), intermediate variables (clinical scores), and target variables (severity, complications, evolution, death). Parameter learning used maximum likelihood estimation with Laplace smoothing. Model performance was evaluated using stratified 10-fold cross-validation and compared against logistic regression, random forest, and support vector machine approaches. Results The Bayesian network model exhibited strong diagnostic performance, with an AUC of 0.95 (95% CI: 0.91–0.99) for death prediction, 0.94 (95% CI: 0.90–0.98) for severe outcomes, and 0.93 (95% CI: 0.89–0.97) for unfavorable progression. Stratified 10-fold cross-validation yielded a mean accuracy of 0.85 ± 0.03 . The model outperformed logistic regression (AUC 0.89), random forest (AUC 0.91), and SVM (AUC 0.87) while maintaining superior interpretability. Sensitivity analysis identified qSOFA score (mutual information: 0.342), SpO 2 levels (0.298), and respiratory distress (0.276) as the most influential variables. Robustness testing showed prediction stability under parameter perturbations of ± 15%, with variation remaining below 8%. Conclusions Bayesian networks constitute a promising tool for COVID-19 outcome prediction in resource-limited settings, combining competitive predictive performance with essential clinical interpretability. The probabilistic approach enables rigorous uncertainty quantification critical for medical decision-making. While the model achieved excellent performance on our cohort, external validation across different populations and SARS-CoV-2 variants is needed to establish broader generalizability.